Presumptions in Argumentation: A Systematic Analysis

Citation preview

PRESUMPTIONS in ARGUMENTATION A Systematic Analysis Petar Bodlović

Presumptions in Argumentation A Systematic Analysis

Petar Bodlović

Presumptions in Argumentation A Systematic Analysis Copyright © Petar Bodlović, Groningen, The Netherlands, 2022 Coverimage: René Magritte, Décalcomanie (1966) Layout & Design: Publiss | www.publiss.nl Printing: Ridderprint | www.ridderprint.nl

Presumptions in argumentation A systematic analysis

PhD thesis

to obtain the degree of PhD at the University of Groningen on the authority of the Rector Magnificus Prof. C. Wijmenga and in accordance with the decision by the College of Deans. This thesis will be defended in public on Monday 21 February 2022 at 12.45 hours

by

Petar Bodlović born on 31 October 1988 in Dubrovnik, Croatia

Supervisors Prof. J.A. van Laar Prof. B.P. Kooi

Assessment Committee Prof. C. Dutilh Novaes Prof. H.B. Verheij Prof. L. Bermejo-Luque

I know the burden’s heavy As you wheel it through the night Some people say it’s empty But that don’t mean it’s light. —Leonard Cohen, A Street

Contents Acknowledgments Introduction 1 Research Subject 2 Research Aims 2.1 Research Questions 2.2 Theoretical Objectives 3 Research Methodology 3.1 Philosophical Tradition 3.2 Theoretical Background 3.3 Research Type and Presentation Tools 4 Research Structure and Results

1 7 9 11 12 13 14 14 14 16 17

Chapter 1: Presumptions, and How They Relate to Arguments from Ignorance 1 Introduction 2 The Standard Dialectical Account: Presumptions as Starting Points 2.1 Legal Presumptions and Initial Dialectical Presumptions 2.2 Foundations of Initial Presumptions: Inferential Reconstruction 2.3 Dialectical Rights and Obligations in Presumption Criticism 3 The Argument from Ignorance: The Standard View 4 Is the Strong Similarity Thesis Tenable? 4.1 The Interpretations of Similarity 4.2 The Structural Perspective: The Nature of Dependency Relations 4.3 The Normative Perspective: The Content of Acceptability Conditions 4.4 SST and Dialectical Rights and Obligations 5 What is the Upshot of the Previous Analysis? 6 Conclusion

21 23 26 28 29 31 32 34 34 36 38 41 44 45

Chapter 2: On the Differences Between Practical and Cognitive Presumptions 1 Introduction 2 Practical Presumptions: The Standard View 2.1 The Dialogical Function of the Practical Presumption 2.2 Practical Presumptions and Presumptive Reasoning 2.3 What Can Defeat Practical Presumptions, and When? 3 Cognitive Presumptions, and How They Differ from Practical Presumptions 3.1 Rescher’s Account of Cognitive Presumptions 3.2 What is The Dialogical Function of Cognitive Presumptions? 3.3 Cognitive Presumption and Presumptive Reasoning 3.3.1 Is There a (Comparable) Deliberation Problem? 3.3.2 Does the Cost of Action Depend on Proceeding on an Error? 3.3.3 Does the Cost Depend on the Goals of Distinct Normative Importance? 3.4 What Can Defeat Cognitive Presumptions, and When? 4 Conclusion

49 51 54 55 56 60 63 64 66 69 71 74 75 78 80

Chapter 3: On Presumptions, Burdens of Proof, and Explanations 1 Introduction 2 Common Principles of Asymmetrical Allocation 3 Cognitive and Practical Presumptions 4 Presumptions and Different Conceptions of the Burden of Proof 4.1 Presumptions and the Burden of Reasoning 4.2 Presumptions and the Burden of Arguing 4.3 Cognitive Presumptions and the Burden of Explanation 4.4 Practical Presumptions and the Burden of Explanation 5 Conclusion

85 87 90 93 96 98 99 100 104 110

Chapter 4: On Presumptions, and Asymmetrical Allocations of Burden of Proof 1 Introduction 2 Preliminary Clarifications 2.1 Kinds of Burden of Proof 2.2 Two Meanings of Allocation 3 Presumptions and Binary Asymmetries 3.1 Practical Presumption and LBBA 3.2 Cognitive Presumption and LBBA 3.3 Practical Presumption and HBBA 3.4 Cognitive Presumption and HBBA 4 Presumptions and Degree Asymmetries 4.1 Assessment Criteria Degree Asymmetries 4.2 Evidence Accessibility Degree Asymmetries 4.3 Commitment’s Strictness Degree Asymmetries 5 Conclusion

113 115 118 118 121 122 123 129 132 133 137 137 140 142 145

Chapter 5: On the Strength of Presumptions 1 Introduction 2 Strength in Presumption Theory 2.1 Conceptual Research 2.2 Strength Research 3 What Determines the Strength of a Presumption? 3.1 Contextual Factors 3.1.1 The Dialogue’s Collective Goal 3.1.2 The Dialogue’s Cooperativeness 3.1.3 Risks, Stakes, and Background Information 3.1.4 The Intended Audience 3.2 Justificatory Factors 3.2.1 The Basic Fact 3.2.2 The Presumptive Rule 3.3 Deontic Factors 4 Cognitive Presumptions and the Weight of Rebuttal 5 Conclusion

149 151 153 153 154 155 155 155 157 158 159 160 160 161 167 169 171

Chapter 6: Bringing It All Together: Presumption Models, and Conceptual Consequences 1 Practical Presumption (PP) 1.1 Basic Characterisation of PP 1.2 The Logical Function of PP 1.3 Justification and the Pragmatic Function of PP 1.4 Patterns of Interaction and the Deontic Function of PP 1.5 Dialogical Functions of PP 1.6 Conditions of PP’s Defeat 1.7 The Strength of PP 2 Cognitive Presumptions (CP) 2.1 Basic Characterisation of CP 2.2 The Logical Function of CP 2.3 Justification and the Pragmatic Function of CP 2.4 Patterns of Interaction and the Deontic Function of CP 2.5 Dialogical Functions of CP 2.6 Conditions of CP’s Defeat 2.7 The Strength of CP 3 What Should the Concept of Presumption Amount to? 3.1 The Limitations of Normative and Functional Conceptions of Presumption 3.2 Towards a Pluralistic Conception of Presumption 4 Conclusion

175

Conclusion 1 A Model of Practical Presumption 2 A Model of Cognitive Presumption 3 The Concept of Presumption 4 Guidelines for Further Research

207 210 211 213 213

Summary Samenvatting Sažetak Curriculum Vitae References

217 219 221 223 225

178 178 179 180 183 186 186 188 189 189 190 191 194 197 198 200 200 202 204 205

A

ACKNOWLEDGMENTS

Acknowledgments

Acknowledgments

Acknowledgments

A

For the past four years, I have been conducting research in my favourite academic discipline, on the topic of my choosing, in a very stimulating institutional environment (Faculty of Philosophy—University of Groningen), and thanks to the ‘RUG PhD scholarship’, I was funded while doing it. Since very few people get this kind of opportunity, I felt privileged from day one. Now, as I approach the final days of my doctoral research, I consider myself lucky because the image of a sleep-deprived PhD student who struggles to manage his workload and is in a very lonely place with regard to developing his ideas never applied to me. There are a few people I would like to thank for this. More than anyone else, I’d like to thank my primary supervisor, Jan Albert van Laar. Jan Albert, I have nothing but warm words for you. You taught me so many things: from helpful writing tips (e.g., structuring the paragraph in a reader-friendly way) to more abstract, theoretical dimensions of argumentation. But what I appreciate the most are three ‘time-related’ aspects of your supervision. First, you were there for me the entire time. Whenever I needed help, I could count on getting it: be it on something minor (how to revise my 32nd footnote) or major (how to respond to a demanding journal reviewer). Second, you were always on time: in almost four years of collaboration, you never postponed a single meeting. Most importantly, due to your relaxed and positive personality, I always had a good time during our sessions. Doing philosophy becomes much easier when done in good company, with an occasional outburst of laughter. Thank you for being a thorough and constructive, yet also a dependable and friendly supervisor. Second, many thanks to my other supervisor Barteld Kooi. Barteld, it was great having you on board. Despite coming from a different academic background (formal logic), you were genuinely interested in my research, and my dissertation has profited a lot from your ‘outside the box’ comments. It always fascinated me how easily you engage in high-quality discussions about topics that fall outside your primary scope of expertise. I am grateful for your many suggestions about writing, publishing, and career planning. Third, many thanks to Catarina Dutilh Novaes who supervised the first nine months of my project. Catarina, thank you for encouraging me to reflect more on the practical value of philosophical ideas. You taught me that aside from focusing on precision and logical soundness, I should also consider the practical, real-life implications of arguments. Also, you taught me that in writing a scientific paper, I should know who the intended audience is and keep this in mind the entire time. Today, these essential lessons seem obvious, but I really needed them back in the day. Although this thesis was written at the University of Groningen, my research on presumptions started at the University of Granada (Spain) under Lilian Bermejo-Luque’s supervision. Lilian, it is difficult to put into words just how much your mentorship meant to me. You were the first international expert I collaborated with. You supervised my first serious piece of work on presumptions and introduced me to a broader argumentation 1

Acknowledgments

community. Since applying for a ‘RUG PhD Scholarship’ was your idea, this dissertation wouldn’t have been written without your generous recommendation. Thank you for teaching me that, contrary to a common stereotype, academic work is not just a cold, dispassionate analysis but fighting for one’s beliefs by all rational means. I feel obligated to thank some scholars from a broader academic community. First of all, dr. David Godden and dr. James B. Freeman for their constant encouragement, but primarily for discussing my work at conferences. I thank dr. Bart Verheij for many thoughtful comments on the late version of the manuscript. Going back to the early days at the University of Zagreb, I’d like to express my gratitude to dr. Dragana Sekulić. Dragana, without your undergraduate course on critical thinking, your supervision of my master thesis, and, above all, your enthusiasm and kindness, I wouldn’t be studying argumentation today. I am also grateful to dr. Srećko Kovač for teaching me the foundations of formal logic and writing a recommendation letter for my scholarship application. Finally, I thank the reviewers of Argumentation and Informal logic for their constructive criticisms and Viktor Ivanković for helpful linguistic suggestions. Aside from writing a thesis, pursuing a PhD degree is also about improving various academic skills and expanding one’s knowledge in different domains. To achieve this, I participated in several reading groups at the Department of Theoretical Philosophy. I thank everyone who regularly attended these sessions, e.g., Diego Castro Amenábar, Jan Albert van Laar, Job de Grefte, Leah Henderson, Catarina Dutilh Novaes, Bita Heshmati, Hendrik Siebe, Stipe Pandžić, and Stefan Sleeuw. I am especially thankful to Erik Krabbe, whose expertise, curiosity, and humour were a true inspiration to me. I also attended several training schools and presented at research colloquia, conferences, and workshops. I thank the organizers and participants of these enjoyable events. Finally, I am grateful to the fellow PhD students for making this journey memorable, especially Li-Chin, Kritika, Lieuwe, Yuri, Merel, Grzegorz, César, as well as Javier Ramia, and José Ángel Gascón (from my time in Granada). Three persons deserve a very special mention. First, my partner Karolina Kudlek. Karolina, thank you for being with me through the highs and the lows, providing emotional and intellectual support, and making my life fulfilled. Thank you for all the long walks and movie nights, parties and holiday trips, breakfasts and dinners, serious and less serious conversations. Thank you for discussing philosophical problems with me, even when this meant listening to my long and confusing ramblings. Without your generosity, I would have hardly made it. Second, my Chilean friend and colleague Diego Castro Amenábar. Diego, I admire your openness to the world and your proactive attitude that made hanging out with you so enjoyable. Thanks for all the intelligent discussions, fun times, and excellent beer suggestions. Finally, a big thanks goes to my old friend Stipe Pandžić. Stipe, you were the only familiar face when I first came to Groningen, and your friendship meant a lot to me. Thanks for all the fun we had.

2

Acknowledgments

None of this would be possible without my father Zoran and my mother Zorana, who supported me unconditionally over the years, provided a strong safety net, and enabled me to get higher education. Mama Zorana, pape Zoran, ovaj doktorat je vaš koliko i moj. Sreća je imati ovakve roditelje i hvala na svemu!

A

Finally, aside from thanking people for helping me to get through my PhD project, I must thank my PhD project for helping me to get through COVID-19’s lockdowns, quarantines, and curfews. Dear project, thanks for giving me a sense of purpose when the virus put normal life on pause. In this vein, I am grateful to the musicians whose work gave me comfort during lonely periods of social distancing: Bob Dylan, The Rolling Stones, Leonard Cohen, Iggy Pop, Lou Reed, Neil Young, The Beatles, Eric Clapton, David Bowie, Bruce Springsteen, Mark Knopfler, Pips Chips & Videoclips, Električni orgazam, Partibrejkers, and Eksperimentalni šljunak.

I dedicate this dissertation to everyone who has ever taught me anything.

3

I

INTRODUCTION

Introduction

“You know, you’re a little complicated after all.” “Oh no, … I’m just a whole lot of different simple people.” —F. Scott Fitzgerald, Tender Is the Night

6

Introduction

Introduction The internet meme ‘Change My Mind’ portrays a cheerful man inviting passers-by to discuss highly controversial statements. He is sitting in a public space, behind a desk with a sign that reads, for instance, “Male privilege is a myth / Change My Mind,” “Coronavirus is a hoax / Change My Mind,” “Penguins are panda chickens / Change My Mind,” or “Cereal is a soup / Change My Mind.”

I

What makes the ‘Change My Mind’ meme so popular? In my opinion, the meme is funny because there is something deeply irrational about the situation it depicts and popular because we can easily relate to this irrationality. We are shocked but also entertained by the fact that people believe absurd claims. Also, we are shocked, as well as entertained by the fact that those who advance absurd claims often win debates. It is easy to imagine a confident eccentric proponent using a myriad of creative rhetorical strategies to refute the opponent’s arguments favouring an obviously plausible view, such as “Cereal is not a soup.” As the dialogue progresses, the opponent will likely become frustrated; her arguments will become ill-shaped and hasty until she finally quits the discussion by offering, for instance, the petitio principii: “Cereal is not a soup, because it just isn’t. Debating this was a bad idea to begin with.” Since everyone has lost a debate about a plausible standpoint at least once, the ‘Change My Mind’ meme is popular because it relies on universal experience.1 But why do eccentric proponents win debates? One part of the answer lies in the argumentative competencies of debating parties. Sometimes, the eccentric proponent is simply a better rhetorician: he is (seemingly) more informed about the subject being discussed; he could also be more intelligent, motivated, eloquent, or charismatic than his opponent. An interesting part of the answer, however, lies in accepting a problematic dialectical rule that distributes the parties’ commitments in a questionable way. For instance, the eccentric proponent may win a discussion by avoiding an explicit duty to defend his standpoint while demanding persuasive arguments for the opponent’s uncontroversial view. We can dialectically reconstruct the ‘Change My Mind’ meme along these lines: (a) after advancing an implausible standpoint (“Cereal is a soup”), (b) the eccentric proponent requests persuasive reasons from the opponent (“Change my mind”) while (c) avoiding an explicit commitment to defend his standpoint. Since the proponent can focus solely on challenging and attacking the opponent’s position, by avoiding to take up the burden of proof, he gets the upper hand. When a particular dialectical rule gets us in trouble, we may hope for an elegant solution: substituting the problematic rule with an alternative one. Instead of a rule exempting the eccentric proponent from supporting his standpoint, we may propose a 1  Of course, there may be other reasons for the popularity of the ‘Change my mind’ meme. For instance, we may find it valuable that even our most agreed-upon opinions can be debated and, possibly, proven wrong. However, the realization that things are not set in stone is hardly humorous, and internet memes are usually made to be funny.

7

Introduction

distribution of commitments that appears more just to a reasonable opponent. What exactly should this distribution look like? According to Rescorla (2009a), there are two distinct schools of thought with regard to this question: dialectical egalitarianism and dialectical foundationalism. Dialectical egalitarians claim that whenever the opponent challenges the proponent’s standpoint, the latter is obliged to defend his view (unless he decides to retract it). Otherwise, we face the threat of dogmatism. Among philosophers, dialectical egalitarianism has been advocated by Pyrrhonian skeptics, and more recently by O. Neurath, P. Klein, and M. Rescorla (see Rescorla 2009a, p. 86). Among argumentation theorists, it has been defended, e.g., by pragma-dialecticians: The discussant who is challenged by the other discussant to defend the standpoint … is always obliged to accept this challenge, unless the other discussant is not prepared to accept any shared premises and discussion rules; the discussant remains obliged to defend the standpoint as long as he does not retract it and as long as he has not successfully defended it against the other discussant… (van Eemeren & Grootendorst 2004, p. 139). In the situation depicted by the ‘Change My Mind’ meme, egalitarians propose a symmetrical, impartial distribution of the burden of proof. Both the eccentric proponent and the reasonable opponent must defend their views (if requested by the other party), and nobody gets the upper hand. By contrast, dialectical foundationalists claim that, if the opponent challenges the proponent’s highly plausible standpoint, the proponent is not immediately obliged to respond to the challenge. Otherwise, we might undermine the efficiency of the dialectical procedure (Aijaz et al. 2013, p. 273), risk dialectical regress (Rescher 2006; Rescorla 2009b), and face counterintuitive consequences (Rescorla 2009a). Prominent advocates of dialectical foundationalism are Aristotle, Wittgenstein, W. Sellars, R. Brandom, J. Adler, and A. Leite (see Rescorla 2009a, p. 86), as well as T. Reid (1785), R. Whately (1963), G. E. Moore (2000), N. Rescher (1977), and J. Freeman (2005). In the situation depicted by the ‘Change My Mind’ meme, foundationalists would propose an asymmetrical, dialectically biased allocation of the burden of proof: until or unless the eccentric proponent offers a persuasive argument for “Cereal is a soup,” the opponent is not obliged to defend her commonsensical view. Initially, the reasonable opponent is entitled to only challenge and attack the proponent’s controversial position, so she gets the upper hand. As a result, dialectical foundationalists suggest that there is a set of dialogically privileged propositions, i.e., statements that, once challenged, do not require an immediate (sufficient) defence, but instead place the burden of proof on the challenger. Traditionally, legal scholars, philosophers, and argumentation theorists label these statements as presumptions. Presumptions, or, more precisely, the standard accounts of presumption are the focus of this dissertation. 8

Introduction

1 Research Subject The notion of presumption leads a double life. On the one hand, ordinary people use it in everyday parlance, within loosely regulated conversational contexts. On the other hand, scholars use it in their theories, within strict, abstract, and tightly regulated frameworks. In natural language, ‘presumption’ has several meanings and I present the most relevant ones below.

I

1. A presumption is an “act of presuming or accepting as true” (The American Heritage Dictionary of the English Language). 2. A presumption is “what the presumer really believes, till it is disproved, to be true” (A Dictionary of Modern English Usage). 3. A presumption is connected to having “a strong reason for believing something to be so” (Merriam-Webster.com Dictionary). It is a “belief based on reasonable evidence” (The American Heritage Dictionary of the English Language), or a “belief that something is true because it is likely, although not certain” (Cambridge Online English Dictionary). 4. A presumption is an “act of believing that something is true without having any proof” (Cambridge Advanced Learner’s Dictionary), or “something believed to be so but not proved” (Merriam-Webster.com Dictionary). 5. A presumption is a “conclusion derived from a particular set of facts based on law, rather than probable reasoning” (The American Heritage Dictionary of the English Language). Obviously, the colloquial concept of presumption is incoherent. Although it always includes ‘accepting a proposition as true,’ everything else seems unspecified. For instance, is presumption an act of accepting the proposition (meanings 1 and 4), or is it the proposition that is being accepted (meaning 2)? If presumption is a proposition, is it always associated with a belief (meanings 2 and 3), or can it represent some other kind of conclusion (meaning 5)? Does reasonable evidence always support a presumption (meaning 3), potentially support a presumption (meaning 5), or is it that it never supports a presumption (meaning 4)? Since the everyday meanings of presumption denote incompatible phenomena, there is a reasonable concern that “the ordinary-language analysis of the notion of presumption … will not get us very far” (Ullmann-Margalit 1983, p. 144). In order to narrow down the research subject and avoid arbitrariness, I rely on the typical theoretical characterizations of presumption instead of everyday meanings. Luckily, presumptions have an exciting academic life. Despite drawing inspiration from different sources, argumentation theorists typically portray presumptions as propositions that impose the burden of proof on those who challenge them. This characterization—primarily inspired by the judicial traditions and introduced to argumentation studies by Richard Whately—is shared by standard accounts of presumption. 9

Introduction

‘Presumption’ in favour of any supposition … implies that it must stand good till some sufficient reason is adduced against it; in short, that the Burden of proof lies on the side of him who would dispute it. (Whately 1963, p. 112) [The presumption rules] also have a procedural aspect, that of casting a burden of producing counterevidence upon the presumption opponent. (Ullmann-Margalit 1983, p. 150) [W]here a proposition has the status of a presumption, then the burden of proof lies with anyone who refuses to concede it. (Pinto 2001, p. 4) [T]he opposite of burden of proof is presumption. (Freeman 2005, p. ix) The existence of a presumption places upon anyone who would deny its application in a particular case the burden of producing good reasons for doing so in the circumstances at hand. (Rescher 2006, p. 15) Presumptions are most usefully represented as modal operators modifying the status of a claim in an argument—indicating that the burden of proof lies with anyone who would reject the claim. (Godden & Walton 2007, p. 334) Roughly, two paradigmatic kinds of foundations generate presumptive status. First, the proponent may be exempted from taking up the burden if his standpoint is epistemically justified. For instance, the proponent should not be immediately obliged to defend “Cereal is not a soup” because, objectively, her view is uncontroversial and wellsupported. Admittedly, discussing conceptual connections between cereal and soup might be stimulating, even entertaining, but the initial burden of proof should be shouldered by those advancing controversial positions. On Rescher’s account, “Cereal is not a soup” would have the status of cognitive presumption since it “represents our most plausible candidate for truth” (2006, p. 71). Various epistemic considerations may generate a proposition’s plausibility, such as personal epistemic sources (“Presumably(c), the cat is in a tree” because I see it there), interpersonal sources (“Presumably(c), the Earth is globeshaped” because scientists say so) (Freeman 2005, p. 41), but also relevant warranting principles, such as simplicity or uniformity (Rescher 2006, p. 40). Regardless of the nature of an epistemic warrant, however, cognitive presumptions remain acceptable until their opponents present persuasive arguments. Second, the proponent may be exempted from taking up the burden of proof due to precaution, i.e., because accepting his standpoint avoids greater harm in deliberation. Suppose we deliberate whether to wear face masks during the COVID-19 pandemic. Our decision depends on whether wearing masks protects other people from COVID-19, and let us assume arguendo that this is an open question. In such circumstances, proceeding on “Face masks protect people from COVID-19” and wearing masks seemingly avoids the greater harm (infecting people with a dangerous disease) than not wearing masks (avoiding a slight inconvenience). Hence, those advocating precaution should not be obliged to successfully defend “Face masks protect people…” to win the practical debate. 10

Introduction

Like “The accused is innocent” in a criminal trial, “Face masks protect people…” has the status of a practical presumption: it remains acceptable until defeated by persuasive arguments (Ullmann-Margalit 1983, Godden 2017). The distinction between cognitive and practical presumptions will play a prominent role in later chapters.

I

Standard accounts of presumption face several objections. For instance, inspired by speech-act theory and philosophy of ordinary language, Kauffeld (2005) objects that traditional accounts are simplistic and rarely accord with argumentative reality. Instead of associating presumptions with epistemic foundations and harm reduction, he relates them to social commitments: “[T]o presume that p is to take that p on the grounds that someone will have made that the case rather than risk criticism, painful regret, reprobation, lose of esteem [sic] or even punishment for failing to do so” (2003, p. 140). Most importantly, it has been argued that, sometimes, presumptions do not concern the burden of proof (see Kauffeld 2003; Bermejo-Luque 2019). These objections are undoubtedly significant for the study of presumption. It is noteworthy, however, that the methodological commitments underlying them are not typical to standard accounts. For instance, the claim that applying the Whatelian notion of presumption to non-legal dialogues is methodologically inadequate is quite fundamental in Kauffeld’s approach (see 1998, 2003, 2005). He is reluctant to study presumptions and burdens of proof in ideal, strictly regulated dialectical models where discussants should obey rules to achieve the dialogue’s ultimate goal (e.g., a resolution of a difference of opinion). By contrast, in line with so-called design theory, he analyses presumptions and probative obligations in real-life, informal dialogues where parties create norms during interaction in order to achieve their own individual goals (for a detailed account of design theory, see Goodwin 2001, 2007). Criticizing some top-down dialectical framework by adopting a bottom-up rhetorical perspective is sometimes legitimate. However, this dissertation takes an alternative approach: in order to evaluate standard accounts internally, on their own merits, it does not challenge their methodological starting points. Rather, it seeks to fully exploit the resources of traditional accounts so as to test their strengths, and not to do away with these accounts prematurely. For the most part, this research refines standard accounts by clarifying claims, reconstructing missing parts, and adding conceptual and typological detail.

2 Research Aims This dissertation investigates standard, Whatelian, judicially inspired accounts of presumption from the viewpoint of argumentation theory. But what exactly are its research questions and theoretical objectives?

11

Introduction

2.1 Research Questions Godden & Walton (2007, p. 314) identify four aspects of presumptions essential for developing a systematic theory: the nature, function, foundation, and force of presumption. Below, I extend their list of essential aspects, and identify questions explored in this dissertation. First, what are presumptions? Are they best modelled as claims (e.g., “Cereal is not a soup”), proposition statuses (e.g., “Presumably, cereal is not a soup”), relations between propositions, discussants, and junctures in the dialogues (see Freeman 2005, pp. 26-33), or speech acts (e.g., “Cereal is not a soup, I presume?”)? Should we treat presumption as a unique speech act resembling both assertion and assumption (Walton 1993, p. 125), an assertive (van Eemeren & Grootendorst 2002, p. 17), a special type of assumption (BermejoLuque 2016, p. 9), or, perhaps, a subclass of supposition (Kauffeld 2003, p. 139)? This dissertation does not investigate speech-act accounts and characterises presumptions as qualified propositions. Occasionally, I comment on the latter views, but the presumption’s most fundamental (ontological) characterization is not the dissertation’s primary subject. Second, what is the logical function of a presumption? How does presumption relate to defeasible reasoning and practical argument? Is it a rule of reasoning (“If X is accused of crime, X is presumed innocent”) (Prakken & Sartor 2006), a conclusion of reasoning (“Presumably(p), John is innocent”) (Walton 2014; Godden 2017), or both (Rescher 2006, pp. 33–35)? This dissertation treats presumptions as conclusions of presumptive reasoning. Third, how does one justify a presumption? If presumptions are conclusions that give a dialectical advantage to their proponents, then they should be supported by good reasons. Is presumptive reasoning reducible to the argument from ignorance (see Chapter 1)? Is the presumptive rule justified because using it in various settings avoids greater harm? What values underlie presumptive reasoning, and how are they connected to reducing harm? In Chapter 2, I discuss these questions by developing the schemes of practical and cognitive presumptive reasoning. Fourth, how do presumptions influence interaction in argumentative dialogue? Are the opponents of a presumption truly obliged to defend their rejections? Should they present arguments, or are they also allowed to offer explanations? And what about the proponents of a presumption? Are they completely exempted from taking up the burden of proof, or should they provide arguments despite their corresponding commitment being somehow less demanding? Finally, do practical and cognitive presumptions differ in previous respects? The questions related to presumption’s deontic function are explored in Chapters 3 and 4. Fifth, what are the dialogical functions of presumptions? If they avoid greater harm and shift the burden of proof, how does this benefit the dialectical procedure? Do presumptions prevent (infinite) dialectical regress, or do they enable dialectical progress when evidential resources are insufficient? Moreover, what is their ultimate contextual function? For instance, if presumptions enable dialectical progress, do they contribute to the epistemic goal of acquiring justified belief, or to the non-epistemic goal of making 12

Introduction

the right decision? In Chapter 2, I take into account the differences between practical and cognitive presumptions and address these questions. Sixth, how are presumptions defeated? If they stand until proven otherwise, what exactly are the opponent’s attacking options? Here, we should distinguish two relevant sub-questions. First, what types of evidence can defeat presumptions, in principle? Second, in which dialogical circumstances can particular types of evidence defeat a presumption? In Chapter 2, I tackle these questions by focusing on the effectiveness of the so-called undercutting defeater (see Pollock 1987).

I

Seventh, what determines the strength of a presumption? Sometimes, different presumptions have different strengths. For instance, the presumption of innocence is stronger than the legal presumption of sanity (“If X is accused of crime, then, presumably, X was sane when the crime was committed”). Also, the strength of a particular presumption may vary over the course of a dialogue. Although strength is obviously related to a presumption’s justification, it is typically measured by the weight of the burden: the more difficult it is for the opponent to defeat a presumption, the stronger the presumption (see Ullmann-Margalit 1983, p. 152; Rescher, 2006, p. 18; Godden & Walton, 2007, pp. 337338). In Chapter 5, I argue that the factors affecting a presumption’s strength are not always correlated. Cognitive presumptions may be weakened due to weaker justification, although the opponent’s burden of rebuttal does not become any lighter.

2.2 Theoretical Objectives What do I hope to achieve by discussing these questions? What is the dissertation’s ultimate purpose? In my view, the current research has three theoretical objectives. First, its conceptual objective is to analyse concepts essential to standard accounts of presumption, such as ‘argument from ignorance,’ ‘the deliberation problem,’ ‘deliberation pressure,’ ‘evidential uncertainty,’ ‘utility imbalance,’ ‘burden of proof,’ ‘asymmetrical allocation,’ ‘strength,’ etc. From a conceptual viewpoint, ultimately, the research aims to specify a more exact meaning of ‘presumption,’ one which is maximally useful for the further development of argumentation theory. Second, the dissertation has an obvious, dialectical objective: a detailed investigation of the dialogical applications of presumptions. Conceptual clarifications are only a precondition for answering questions such as: Which patterns of interaction do presumptions generate? Which dialogical problems do they solve? How do they arise, and how are they defeated? The ultimate dialectical objective is to contribute to a systematic study of presumptions in argumentative dialogues. Third, the greater part of the research is concerned with the typological objective: exploring the relationship between practical and cognitive presumptions by using conceptual clarifications and dialectical characterizations. Do practical and cognitive presumptions generate identical patterns of interaction? Do they solve comparable dialogical problems? Are they justified on similar grounds, and defeated by the same kinds

13

Introduction

of evidence in comparable dialogical circumstances? Since the relationships between presumptions still lack detailed investigation, this research’s typological dimension fills an obvious gap in the existing literature.

3 Research Methodology After presenting the dissertation’s main subject, research questions, and objectives, I now turn to explaining the research methodology. I distinguish four interconnected methodological aspects: philosophical tradition, theoretical background, research type, and presentation tools.

3.1 Philosophical Tradition The dissertation belongs to the tradition of analytic philosophy. To be sure, the demarcation lines between philosophical traditions are sometimes blurry, but analytic philosophers typically rely on arguments and thorough analysis, proceed in a piecemeal fashion, take into account scientific findings, and use clear, precise language (Glock 2008, pp. 152153). Accordingly, to a significant extent, this research is inspired by the following motto (attributed to the Erlangen School of dialectics): [W]hat one needs, for the advancement of philosophy and the humanities, is not a plethora of brilliant ideas, but rather a “discipline of thought and speech” that will permit us to get rid of speaking at cross-purposes in interminable monologues and to make a new start in a reasonable dialogue (van Eemeren et al. 2014, p. 310) .

3.2 Theoretical Background As far as theoretical background is concerned, the dissertation is a contribution to argumentation theory. However, contemporary argumentation theory is heterogeneous: scholars investigate different subjects (reasoning vs. dialogues) at different levels of abstraction (analysing a parliamentary debate vs. developing a formal dialectical system); they focus on distinct dimensions (epistemic vs. dialectical vs. rhetorical), and conduct different kinds of research (descriptive vs. normative); finally, theorists often start from diametrically opposed methodological perspectives (bottom-up vs. top-down). Bearing these distinctions in mind, I study presumptions normatively (while considering inferential, epistemic, and dialectical dimensions), from a top-down perspective, in a moderately abstract way, and in the context of persuasion dialogue. Explaining the dialectical context within which presumptions operate is especially important for understanding the theoretical framework underlying this study. Presumptions are naturally analysed in the context of persuasion dialogue or critical discussion (due to their explicit connection to the burden of proof). Parties in persuasion dialogues start from “conflicting points of view” (Walton & Krabbe 1995, p. 66) and exchange arguments to “resolve differences of opinions” (van Eemeren & Grootendorst 2004, p. 35). In the 14

Introduction

pragma-dialectical model of critical discussion, there are four ideal stages of the resolution procedure. This four-part division will prove useful for comparing presumptions and understanding distinct kinds of burden of proof (see chapters 1, 2, and 4). In the confrontation stage, parties disagree on a particular issue. The objective is to clarify their positions and determine the type of disagreement: whether the proponent’s standpoint “runs up against doubt or contradiction” (van Eemeren & Grootendorst 2004, p. 60). If the opponent doubts the standpoint (“I am not convinced that cereal is a soup”), the difference in opinion is non-mixed. But if she puts forward a contradictory standpoint (“Cereal is not a soup”), we are faced with a “mixed dispute” (van Eemeren & Houtlosser 2002, p. 18). I will mostly analyse presumptions in mixed disputes where Proponent 2 rejects Proponent 1’s presumption by advancing a contradictory standpoint.

I

Then, in the opening stage, parties accept suitable roles. The party advancing a standpoint must assume the role of the proponent (protagonist) and defend her standpoint. The party doubting a standpoint must assume the role of the opponent (antagonist) and critically test the proponent’s contributions (van Eemeren & Grootendorst 2004, p. 60). However, the central objective is establishing the rules of the dialogue (e.g., “One should not use irrelevant reasons”) and identifying premises conceded by both discussants (e.g., “Cereal and soup are liquid dishes”). In short, parties must “find out how much relevant common ground they share (as to the discussion format, background knowledge, values, and so on) in order to be able to determine whether their procedural and substantive ‘zone of agreement’ is sufficiently broad to conduct a fruitful discussion” (van Eemeren & Grootendorst 2004, p. 60).2 In the argumentation stage, which lies at the center of the dialogue, parties act in accordance with their roles and the discussion—understood in the narrower sense—takes place. To overcome the doubt expressed by the opponent, the proponent defends her standpoint (e.g., “Cereal is a soup because it is a liquid dish”), and the opponent evaluates the quality of the defence. If the opponent remains sceptical, she reacts critically (“Are you implying that every liquid dish is a soup?”) and forces the proponent to provide further support (see van Eemeren & Grootendorst 2004, p. 61). Thus, the main dialectical objective in the argumentation stage is “to test the tenability of the standpoints … starting from the point of departure established in the opening stage” (van Eemeren & Houtlosser 2002, p. 24). In the concluding stage, parties “establish the result of the critical testing procedure” (van Eemeren & Houtlosser 2002, p. 25). If the proponent responds in a satisfactory way to all the critical reactions, the opponent must retract her doubt and the standpoint is successfully defended. If the proponent fails to respond to all the critical reactions, the sceptical doubt remains, and the proponent loses. To fully complete the resolution procedure, parties must agree on the discussion’s outcome (see van Eemeren & Grootendorst 2004, pp. 61-62; van Eemeren et al. 2014, p. 530). 2  The discussion will hardly get off the ground if parties disagree about the rules of the game. Also, the proponent will hardly persuade the opponent if the latter does not concede her argument’s premises.

15

Introduction

Persuasion dialogues differ with respect to “the nature of the subject matter discussed” (Walton & Krabbe 1995, p. 70). Accordingly, cognitive presumptions will be explored within epistemic dialogues about doxastic states (“Is the Earth globe-shaped or not?”, “Is the cat in the tree, or must we suspend judgment?”), and practical presumptions will be explored within non-epistemic dialogues about actions (“Shall we take the umbrella or not?”, “Shall we attend the party or not?”). I will call a persuasion dialogue about doxastic states an inquiry, and a persuasion dialogue about actions a deliberation (for different, and more typical terminological choices, see Walton & Krabbe 1995; Walton 1998, Atkinson et al. 2013).

3.3 Research Type and Presentation Tools Is this research best described as conceptual analysis, or perhaps conceptual engineering? Is it associated with the traditional armchair approach, or based on empirical data? If understood in the broadest sense as “specifying the meanings of words and sentences, the concepts involved in an assertion” (Daly 2010, p. 218), this research surely includes conceptual analysis. It investigates statements like “Presumptions are closely similar to arguments from ignorance,” “Presumptions avoid greater harm,” and “Presumptions asymmetrically allocate the burden of proof” by scrutinizing the concepts involved. More specifically, however, conceptual analysis is the process of identifying the concept’s necessary and sufficient conditions by testing intuitions in hypothetical cases (Kornblith 2016, p. 152). For instance, what can we intuitively say about knowledge and justice? Does S’s true belief “X is a real barn” represent knowledge if X happens to be surrounded by fake barns (Goldman 1976)? Is returning a borrowed weapon just if the weapon’s owner happens to be a madman who can easily harm somebody (Plato 2000)? Conceptual analysis usually deals with non-technical expressions. So, for the most part, the analyst must specify what concept ordinary people (should) have in mind when using a common expression (Kornblith 2016). Consequently, it wouldn’t be entirely correct to describe this research as conceptual analysis. As I already stated, instead of analysing presumptions as used in everyday (sometimes non-argumentative) conversations, the dissertation explores presumptions as portrayed by standard argumentative models. Also, aside from analysing how standard models construct presumptions, this research is a normative study of how models should construct presumptions. Hence, the research to a significant extent exemplifies Carnapian explication (see Carnap 1947; Dutilh Novaes & Reck 2017) or conceptual engineering understood as “the process of designing, implementing, and evaluating concepts” (Chalmers 2020, p. 2). In most cases, my analyses in the following chapters will start from (1) accepting some concept designed by theorists, proceed by (2) implementing this concept to different stages in persuasion dialogues, and end by (3) evaluating implementation results. If the concept’s implementation is problematic, conceptual engineering will make the revised concept less ambiguous, and 16

Introduction

more exact. However, the final product of conceptual engineering—unlike the final product of typical conceptual analysis—is more of a cognitive invention than a cognitive discovery. All this indicates that my research is not empirical, or that it is empirical inasmuch as it uses mundane, simple examples to explain theoretical points (e.g., “How does presumption work in deliberation about bringing an umbrella on a cloudy day?”).

I

Finally, as far as presentation tools are concerned, I put arguments in the standard form (Chapter 2), or represent their structures by diagrams (Chapter 6). I explain patterns of interaction by using dialogue profiles, and classifications by using diagrams and tables.

4 Research Structure and Results What is the dissertation’s structure, and what are its main results? In what follows, I summarize the ground that the upcoming chapters cover and the outcomes that they produce. Chapter 1 is a linguistically improved version of the article “Presumptions, and How They Relate to Arguments from Ignorance,” published in the journal Argumentation (2019, Vol. 33,  pp. 579–604). It explores Douglas Walton’s thesis that arguments from ignorance and presumptive reasoning are almost indistinguishable. I put this ‘strong similarity thesis’ to the test from three different perspectives: (1) inferential structures associated with ad ignorantiam and presumptive reasoning; (2) evidence that potentially defeats ad ignorantiam and presumptive reasoning; and, finally, (3) the attacking rights and obligations assumed by the opponents of ad ignorantiam and presumptive reasoning. The chapter concludes that the strong similarity thesis is untenable. Result 1. Cognitive (initial) presumptions, unlike practical ones, are not closely similar to arguments from ignorance. Chapter 2 corresponds to the article “On the Differences Between Practical and Cognitive Presumptions,” published in the journal Argumentation (2021, Vol. 35, pp. 287-320). First, it explores Nicholas Rescher’s contention that both types of presumption are made to avoid greater harm in circumstances of epistemic uncertainty. I reconstruct schemes of practical and cognitive reasoning, and show that Rescher’s contention requires qualifications. The chapter also investigates the dialogical functions of presumption, as well as the conditions of its defeat. Result 2. Practical and cognitive presumptions have relevantly different justifications. Unlike the scheme for practical reasoning, the scheme for cognitive reasoning does not include premises related to genuine uncertainty and deliberation pressure. Thus, the two types of presumptions have different pragmatic functions: they are associated with distinct policies of harm reduction. Result 3. Practical and cognitive presumptions perform different dialogical functions: typically, the former enable dialectical progress when evidence is uncertain, while the latter enable a different kind of progress, or prevents dialogical regress. 17

Introduction

Result 4. Presumptions are defeated under different conditions: sometimes, in comparable dialogical settings, the opponent may successfully undercut a cognitive, but not a practical presumption. Chapter 3 is a linguistically improved version of the article “On Presumptions, Burdens of Proof, and Explanations,” published in the journal Informal logic (2020, Vol. 40, No. 2, pp. 255–294). It investigates the deontic function of presumption, i.e., whether presumptions, once rejected, impose the burden of proof (understood as an obligation to present an argument) on their opponents. The chapter shows that practical and cognitive presumptions have different deontic functions. Result 5. After rejecting a cognitive presumption at tn, the opponent does not incur the burden of proof at tn+1: she is allowed to offer an argument, but also an explanation. By contrast, after rejecting a practical presumption at tn, the opponent is strongly recommended to argue at tn+1. Chapter 43 further explores presumption’s deontic function. When a presumption p asymmetrically allocates the burden of proof, what exactly does ‘asymmetrical allocation’ mean? Is the proponent of p completely exempted from the obligation to present arguments (binary asymmetry)? Or is she committed to provide arguments, but her burden is, in some sense, lighter than the opponent’s burden (degree asymmetry)? Result 6. Practical and cognitive presumptions often entail the same binary asymmetry, but in different dialogical settings. For instance, if the opponent successfully defeats p at tn, the proponent is exempted from supporting the cognitive presumption p, but not the practical presumption p at tn+1. Result 7. Presumptions can entail different degree asymmetries, and allocate distinct burdens of proof. For instance, the opponent of a practical presumption should meet the higher standard of proof to win the debate—she carries the burden of persuasion. By contrast, the opponent of a cognitive presumption must fulfil the burden of successful defence to win the debate, but this burden does not entail the higher proof standard. Chapter 5 is almost entirely revised version of the article “What determines the strength of practical and cognitive presumptions?” published in Proceedings of the ninth conference of the International Society for the Study of Argumentation (2019, eds. B. Garssen, D. Godden, G. R. Mitchell, and J. H. M. Wagemans, pp. 106–117).4 On standard accounts, the strength of a practical presumption is related to the rebuttal’s weight: if a presumption p is weaker (stronger), proving non-p is less (more) demanding. Chapter 5 explores one limitation of this view. 3 

Submitted to the journal Informal logic in June 2021.

4 

In its present, extended, and almost entirely revised version, Chapter 5 is submitted to the journal Pragmatics &

Cognition in September 2021.

18

Introduction

Result 8. Determining the strength of a presumption in terms of the rebuttal’s weight is not applicable to cognitive presumptions. A cognitive presumption p may become weaker even if the difficulty of proving non-p remains unchanged. In Chapter 6, I bring all the pieces together and present models of practical and cognitive presumptions in systematic fashion. The chapter shows that even when standard accounts of presumption are internally scrutinized, they require clarifications and revisions. Also, it reveals that the two types of presumptions differ in many fundamental respects, thereby raising concerns regarding the conceptual coherency of presumption. Is the situation in theoretical language as confusing as in ordinary language? Is there some feature holding practical and cognitive presumptions together? Is this feature theoretically interesting? In Chapter 6 I briefly comment on these issues.

I

In Conclusion, I summarize the main results and provide guidelines for future research.

19

1

CHAPTER 1 Presumptions, and How They Relate to Arguments from Ignorance

CHAPTER 1

Those who jump to conclusions may go wrong. —Sophocles, Oedipus the King

22

Presumptions, and How They Relate to Arguments from Ignorance

Presumptions, and How They Relate to Arguments from Ignorance Abstract: By illustrating the argument from ignorance in terms of the presumption of innocence, many textbooks in argumentation theory suggest that some arguments from ignorance might share essential features with some types of presumptive reasoning. The stronger version of this view, which states that arguments from ignorance and presumptive reasoning are almost indistinguishable, is occasionally proposed by Douglas Walton. This chapter explores the features and limits of the strong view and argues that initial presumptions and arguments from ignorance are not closely related. There are three main reasons. First, the argument from ignorance, unlike typical presumptive reasoning, is a negative kind of inference. Second, the typical initial presumption is sensitive to a broader set of defeaters and thus assumes a higher (negative) standard of acceptability. Third, in dialectical terms, initial presumptions and arguments from ignorance bring about different attacking rights and obligations. I conclude that the Waltonian intuition is unsupported or, at best, limited only to practical presumptions and practical arguments from ignorance.

1

Keywords: argument from ignorance, burden of proof, defeaters, Douglas Walton, initial presumption, negative evidence, presumptive reasoning, starting points.

1 Introduction Imagine two tourists, Smith and Jones, losing their bearings in a new city. Smith asks a random person for directions. This random person (call her Jill), who appears to live there, tells him that the shortest way to reach the city center is to take the second street to the right. The circumstances are standard—there is nothing suspicious about Jill or her directions, and Smith finds it reasonable to believe that she is telling the truth. “It’s the second street to the right then” says Smith, Jones agrees with him, and they start walking. However, a few minutes later, Jones turns to Smith and asks him: “What makes us so sure that those directions were correct?” How should Smith respond? Is he immediately obliged to provide (additional) reasons for trusting Jill? Intuitively, he is not. Since Jones unexpectedly retracts his initial commitment (“The shortest way to reach the city center…”), which he based on transparent and uncontroversial grounds, it is him, not Smith, who seems obliged to provide reasons. He could say, for instance: “Don’t you see that we are heading for the suburbs? Jill was probably confused,” or “I read in the travel guide that the locals here deceive tourists for fun.” However, if Jones fails to provide convincing defeaters, Smith’s claim stands. In technical terms, the proposition p (“The shortest way to reach the city center...”) has ‘presumptive’ status: (1) p asymmetrically distributes dialectical obligations, and (2) p is taken as acceptable until or unless the opponent can prove that it is unacceptable. 23

CHAPTER 1

It may seem that this characterization of presumptive status has the structure of an ‘argument from ignorance.’ There are at least two reasons to support this intuition. First, like presumptions, arguments from ignorance are often characterised in terms of shifting (reversing) the burden of proof (see Walton 2014, p. 2; pp. 190-200). Second, the conclusion of an argument from ignorance, similarly to a presumption, is supposed to stand precisely because we are not aware of sufficient reasons to the contrary (Krabbe 1995). On the face of it, the two appear theoretically similar. In argumentation theory, the similarity has been repeatedly stressed by Douglas Walton (e.g., 1993, 1996a, 2001, 2008a), a leading author on both presumptions and arguments from ignorance. In his opinion, the argument from ignorance can explain the central features of presumptive reasoning. Not only is it right to say that the argumentum ad ignorantiam is closely linked to presumptive reasoning and burden of proof. You could even say that the very structure of the argumentum ad ignorantiam is an expression of how presumptive reasoning and burden of proof can function correctly in argumentation to shift a presumption to the other side in a dialogue. (Walton 1992, p. 386; 1996a, p. 125) However, Walton contends that the relation also holds in the opposite direction, i.e., that presumptive reasoning can explain some central features of the argument from ignorance. He defines the argument from ignorance as “a negative type of knowledgebased or presumptive reasoning” (1999, p. 434) and contends that arguments from ignorance “can be quite reasonable presumptive arguments” (2001, p. 157). Various logicians, rhetoricians, and argumentation scholars acknowledge the latter contention since most presentations of the (fallacy of) argumentum ad ignorantiam include a reference to the famous presumption of innocence.5 Given the usual tendency of illustrating the (practical) argument from ignorance in terms of paradigmatic (practical) presumption, one may wonder whether all types of arguments from ignorance can be explained, in relevant respects, by referrence to presumptions and presumptive reasoning. Given this explanatory interdependence, it is quite understandable for Walton to conclude that presumptive reasoning and the argument from ignorance are “so closely entangled that it is hard to distinguish one from the other” (2008a, p. 50). Some scholars seem to share this view. For instance, while discussing the relationship between inquiry and the argument from ignorance, Cummings (2002, pp. 120-1) adopts Walton’s characterization. In a recent paper on the argument from ignorance, Tuzet (2015, p. 46) expresses a rather Waltonian intuition that presumptive arguments (made in ignorance) might “collapse into” arguments from ignorance. Finally, Walton assesses that Alfred Sidgwick’s traditional account of presumption “might amount to nothing more than an argumentum ad ignorantiam” (2014, p. 15). 5 

Usually, this reference seeks to show that arguments from ignorance are not always fallacious (see, e.g., Hamblin

1970; Tindale 2007; van Eemeren, Garssen, & Meuffels 2009; Copi, Cohen, & McMahon 2016, etc.).

24

Presumptions, and How They Relate to Arguments from Ignorance

Are presumption and the argument from ignorance so strongly similar that the former explains the latter and vice versa, as suspected above? To advance a clear-headed theory of presumption, we need to resolve this issue, which is the main aim of this chapter. In it, I explain the various senses in which the two might be considered similar, and argue that the Waltonian intuition is unsupported—the strong similarity thesis (SST) about presumptions and arguments from ignorance is untenable.6 However, it is important to note that the present analysis focuses exclusively on the epistemic versions of presumptions and arguments from ignorance: the chapter explores whether the ‘initial’ dialectical presumption, understood as an epistemic starting point of the dialogue, is strongly similar to the conclusion of an epistemic argument from ignorance. As my introductory remarks indicate, the theoretical similarity appears clear on its face. For this reason, I analyse their relationship in greater depth.

1

The chapter unfolds in five parts. First, I outline a standard dialectical approach to initial presumption (Section 2). Second, I present a ‘textbook version’ of the argument from ignorance (Section 3). Third, I test the purported similarity from two interrelated but distinct perspectives: the structural perspective (Section 4.2) and the normative perspective (Section 4.3). From the structural perspective, the analysis shows that presumptive reasoning and the argument from ignorance “depend” on the absence of defeaters in structurally different ways. From the normative perspective, the analysis shows that presumptive reasoning typically allows the opponent to have more defeating options. As a result, a broader set of negative conditions determines the acceptability of an initial presumption. The analysis also indicates that presumptions and arguments from ignorance do not assign dialectical obligations in a similar way (Section 4.4). In the final section, I explain the importance of the present analysis and sketch some potential implications of its results (Section 5). In general terms, this chapter shows that initial presumptions and the epistemic argument from ignorance merit distinct treatment. Not only do they have different foundations, allow different defeating strategies, and impose different dialectical obligations, but they can also typically be found in different stages of argumentative discussion. While initial presumptions standardly occur in the ‘opening stage,’ arguments from ignorance emerge in the ‘argumentation’ or, perhaps, ‘concluding’ stages of a discussion.7 Taking the strong similarity for granted would conceal these differences, thereby obstructing, rather than stimulating a clear-headed study of presumption. 6  I understand ‘strong similarity’ as a conjunction of two entailments—the ignorance-presumption entailment (IPE) and the presumption-ignorance entailment (PIE)—and argue that both IPE and PIE are untenable (although not always from the same analytical perspective). This version of the SST is certainly inspired by Walton’s remarks, but it is difficult to assess whether it fully captures his view. At least two factors make this assessment difficult—first, Walton’s account of similarity lacks detailed elaboration and, second, his notion of presumption has changed over the years (for a survey, see Bodlović 2017). Hence, the present analysis does not directly focus on criticizing Walton’s original view. Rather, it focuses on testing a version of the SST which, I believe, is the most plausible. 7 

Distinguishing between the opening, argumentation, and concluding stages, primarily associated with the pragma-

dialectical approach to argumentation, is common in argumentation theory (for details, see van Eemeren & Grootendorst 2004, pp. 57-62). Interestingly, pragma-dialecticians also take presumptions to be material starting points of dialogue (see van Eemeren & Houtlosser 2002; Houtlosser 2003).

25

CHAPTER 1

2 The Standard Dialectical Account: Presumptions as Starting Points Studies on presumption in argumentation theory involve different theoretical goals, sources of inspiration, and, accordingly, different methodologies. The more traditional approach is inspired by the judicial (and more technical) meaning of presumption. It rests on legal traditions and explores the relationship between presumptions and obligations in rule-regulated (idealized) dialogues (Kauffeld 2003). The more recent approach is inspired by a non-technical, ‘plain,’ or ‘common’ meaning of presumption. It rests on the concept of presumption used by people in day-to-day communication and explores the relationship between presumptions and obligations by analysing different kinds of speech acts (such as accusing, proposing, and advising) (see Kauffeld 1998, 2003, 2005). In this chapter, I present and analyse presumptions in line with the traditional dialectical approach, inspired by a more technical meaning of presumption. The standard dialectical approach constructs presumptions as propositions that need not be argued for once the opponents challenge them. Since the proponent’s obligation to defend a proposition (if requested by the other party) follows from one of the most fundamental rules of reasonable dialogue (see Rescher 1977, pp. 29-30; van Eemeren & Grootendorst 2004, pp. 138-141), presumptions belong to a set of dialectically privileged propositions. They lift the burden of proof from the proponent and assign the dialectical obligation to the opponent.8 Different explanations of this privilege provide grounds for distinguishing different kinds of dialectical presumptions. For instance, proposition i is dialectically privileged because the proponent is not obliged to provide sufficient reasons for i. Suppose that i has ‘common ground’ status—either the interlocutors have explicitly agreed that i is acceptable or the proponent can reasonably expect that the other party accepts i since, in normal circumstances, i is uncontroversial and accepting it is considered reasonable (e.g., “Smoking is unhealthy”). In such cases, the proponent is not immediately obliged to argue for i if requested to do so. It is important to emphasize that i usually acquires common ground status precisely because the interlocutors are (or should be) aware of sufficient evidence in i’s favour. I call the presumptions of the latter kind ‘initial dialectical presumptions.’9 Initial presumptions are material starting points of a dialogue, and they stand until overturned by a sufficiently strong defeater (see Rescher 2006, p. 25; van Laar 8 

Rescorla (2009a) calls this view dialogical foundationalism. He labels the opposing view—doubting the existence

of privileged propositions in dialogue—dialogical egalitarianism. For details regarding the relationship between the two approaches, see also Rhode (2017). 9 

In many respects, the concept of initial (or shared) epistemic presumption is closely connected to Rescher’s (2006)

notion of “cognitive presumption.” However, our concept of initial presumption might be less restrictive. As presented in this chapter, initial presumptions can arise from both doxastic states (by means of presumptive inference) and non-doxastic states. By contrast, Rescher seems to suggest that cognitive presumptions arise (only) from non-doxastic states and are not inferred from some pre-established premises (2006, pp. 22-23).

26

Presumptions, and How They Relate to Arguments from Ignorance

& Krabbe 2013, p. 210; Lewiński 2017, pp. 604-606). The case of Smith and Jones asking for directions serves as a good example here.10 By contrast, some presumption p can be dialectically privileged because the proponent is obliged neither to provide nor to have sufficient evidence for p.11 The interlocutors are not aware of sufficient evidence for p, but the proponent nevertheless presumes p for the sake of making progress in deliberation. For instance, UllmannMargalit (1983) and Godden (2017) view presumptions as claims that are considered acceptable for non-epistemic reasons in the face of evidential uncertainty and deliberative pressure. They hold that, if (1) the evidence does not sufficiently discriminate between p and non-p, (2) our decision depends on whether p is true, (3) there is a pressure to make a decision, and (4) proceeding on p is more practically justified than proceeding on non-p (e.g., it promotes social safety), then we may presume and act upon p until the opponent provides a sufficiently strong defeater. In these special circumstances, the proponent of p does not shoulder the burden of proof.

1

This chapter focuses on the initial dialectical presumptions. I start from the features of initial (cognitive) presumptions as epistemic starting points of dialogue and explore the similarity between them and the ‘textbook’ argument from ignorance.12 Hence, the conclusions of the analysis should not be uncritically generalized to the so-called ‘practical dialectical presumptions,’ as understood by Ullmann-Margalit, Godden, and, occasionally, Walton (2008b, 2014). Bearing this clarification in mind, let us explore the nature of presumption in more detail. 10 

Presumptions are propositions that are ‘taken as true until defeated.’ It is important to note, however, that not all

propositions of the latter kind are presumptions. Assumptions, for instance, are also taken as true until defeated, but the relatedness between presumptions and assumptions remains uncertain. For instance, there is controversy whether they are distinct kinds of inferences (e.g., Kauffeld 2003, p. 139, 142; Rescher 2006, p. 21) or, unlike presumptions, “assumptions are not the conclusions of previous inferences” (Godden 2017, p. 499). Scholars also disagree whether presumptions and assumptions are distinct kinds of speech acts (e.g., Walton 1993, p. 138), or the difference runs deeper since presumptions (see Kauffeld 2003, p. 144) or, perhaps, assumptions (see Bermejo-Luque 2016, p. 10) are not speech acts. In general terms, there is no consensus on how to logically conceptualize the difference between the two concepts. Some scholars treat presumptions and assumptions as subclasses of suppositions, i.e., as “neighboring subclasses of a larger genus” (Kauffeld 2003, p. 139), whereas others contend that presumptions are a subclass of assumptions (e.g., Scott Jacobs, according to Kauffeld 2003, p. 139; Rescher 2006, p. 29; Bermejo-Luque 2016, p. 9). However, scholars are in agreement that presumptions and assumptions can be distinguished in terms of ‘strength’—presumptions are ‘taken as true’ on stronger grounds than assumptions and, for this reason, they are “load-bearing (unlike assumptions and hypotheses)” (Godden 2017, p. 508) and have some “probative weight” or “bite” (Rescher 2006, p. 20; p. 23). As a result, presumptions are more difficult to defeat. While opponents can (typically) defeat a presumption only by providing convincing reasons, they can defeat an assumption by merely challenging it (see Kauffeld 2003, p. 142) or by simply rejecting it (see Godden & Walton 2007, pp. 326–327). Thus, according to the standard dialectical characterization, presumptions, unlike assumptions, reverse the burden of proof (Walton 1993, p. 138). 11 

For a short discussion on the difference between these two kinds of presumptions in Walton’s theory, see Bodlović

(2017, pp. 521-524). 12 

For the sake of brevity, I will sometimes talk about the similarity between presumptions and arguments from

ignorance. This formulation, however, is neither precise nor entirely accurate. What I in fact do is explore the similarity between typical presumptive reasoning and a typical argument from ignorance, i.e., the similarity between a typical presumption and a typical conclusion of the argument from ignorance.

27

CHAPTER 1

2.1 Legal Presumptions and Initial Dialectical Presumptions In a technical sense, ‘presumption’ is a legal term used to denote various mechanisms of distributing probative obligations.13 A famous legal example is the presumption of innocence, requiring that the accused be presumed innocent until proven guilty. That is, during a criminal trial, the defence is not required to prove the defendant’s innocence, and it is, in that sense, dialogically privileged. By contrast, the prosecution is required to prove the defendant’s guilt (beyond a reasonable doubt) in order to rebut the standing presumption. Argumentation scholars often discuss this rule as a paradigmatic example of presumption (see Macagno & Walton 2012; Godden & Walton 2007; Rescher 2006; Freeman 2005, etc.). Accordingly, it should come as no surprise that initial dialectical presumptions are meant to reconcile the following two intuitions. On the one hand, some propositions should be privileged in a dialogue (i.e., should not shoulder the burden of proof), but, on the other hand, some privileged propositions should be defeasible (i.e., should be retracted once sufficiently strong defeaters arise).14 In our introductory example, both Smith and Jones initially accept p (“The shortest way to reach the city center...”), making p a shared dialogical commitment (a dialogical starting point). As such, it is both privileged and defeasible. It is privileged in the sense that, when Jones challenges p, Smith does not have an obligation to provide arguments (until Jones substantiates his challenge). It is also defeasible in the sense that Smith becomes obliged to retract the presumptive status of p if Jones makes a satisfactory case against it. However, if Jones fails, the presumption stands, and he is obliged to concede it. Most generally, if p is a standing presumption, the opponent can either concede p or attack the acceptability of p – she is obliged to do one or the other (see, e.g., Pinto 2001, pp. 3-4). Suppose now that, at time t, the opponent refuses to concede a presumption (by presenting a challenge). What happens at t1? At t1, the opponent becomes obliged to attack the acceptability of p (while the proponent is exempt from the burden of proof). What exactly does this mean from the opponent’s perspective? What particular dialectical rights underlie her obligation to attack the presumption? What attacking (defeating) strategies is she allowed to use? This appears to be one of the fundamental questions for our purposes. To provide a satisfactory answer, we must discuss the foundations (sources/grounds) of the initial presumptions.

13  A recent analysis of various conceptions of presumption in law was offered by Gama (2017). Surveys regarding legal presumptions can also be found in Walton (2014), Macagno & Walton (2012), and Rescher (2006). 14 

In other words, not every privileged proposition is a ‘fixed concession.’ For a more detailed elaboration of these

intuitions, see Lewiński (2017). On the dialogical properties of ‘fixed concessions’ and ‘presumptions,’ from the perspective of critical reactions they permit, see van Laar & Krabbe (2013).

28

Presumptions, and How They Relate to Arguments from Ignorance

2.2 Foundations of Initial Presumptions: Inferential Reconstruction Although the traditional approach establishes a link between presumptions and a particular distribution of obligations, neither legal nor argumentation scholars agree on the role that presumptions play in this distribution. Some authors define presumption as a rule that authorizes p and, thereby, asymmetrically allocates dialectical obligations (see, e.g., Prakken & Sartor 2006). The legal example of such a rule is “If somebody is accused, then presume that she is innocent until proven guilty.” Others define presumption as the output of a presumptive rule, a conclusion p or, more precisely, a distinctive modal status of proposition p (“[Presumably,] Smith is innocent”) (e.g., Hansen 2003; Walton 2008b, 2014; Godden 2017).15 Defined in this way, a presumption does not allocate the burden of proof (in a strict sense) but rather signals its asymmetrical allocation for the proposition at hand. In this chapter, I follow the latter tradition. The view that we derive presumptions from presumptive rules (and basic facts) will provide a blueprint for explaining the foundations of presumption, which is crucial for comparing presumptions and arguments from ignorance.

1

The usual way of accounting for the foundations of presumption is to treat a presumption as a conclusion of reasoning.16 For instance, Walton (2008b, 2014) interprets ‘presumption’ (presumed fact) as a conclusion of ‘presumptive reasoning,’ which involves a ‘basic fact’ and a ‘presumptive rule.’ Our introductory example illustrates this structure: from (1) the basic fact “Jill testified that the shortest way to reach the city center…” and (2) the presumptive rule “Normally, we should trust other people,” Smith may (3) presume “The shortest way to reach the city center...” (given that the evidence does not indicate anything unusual about Jill’s testimony).17 In Walton’s terms, Smith’s reasoning has the form of a defeasible modus ponens. This kind of inferential reconstruction reveals that presumptions are not stipulations, but rather “conclusions of some prior arguments” (Lewiński 2017, p. 605). Although the proponent is not dialectically obliged to provide prima facie reasons for the initial presumption, she is certainly obliged to have good 15 

The lines dividing these traditions are often blurry. On Rescher’s account, for instance, both presumptive rules

and the results of their application are presumptions: presumptive rules are abstract (generic) presumptions that produce concrete (specific) presumptions upon application (2006, pp. 33-35). 16 

Ullmann-Margalit (1983) studies presumptions (presumed facts) within the deliberative framework and treats them as

conclusions of (practical) reasoning. Rescher emphasizes that presumptions are taken to be true for pragmatic reasons, due to the cost-effective principle (presumptive rule) that they rest upon. Plumer (2017) argues that presumptions are components of arguments – either their assumptions or presuppositions, inherently based on conventional rules. Although his approach significantly differs from the legally inspired dialectical approach to presumptions, Kauffeld also treats presumptions as conclusions of inferences (see Kauffeld 2003). Bermejo-Luque (2016) holds that presumptions are not inherently inferential, but can be used as components of inferences, both as premises and conclusions. Finally, Freeman (2005) contends that presumptions are (typically) based on non-inferential epistemic sources and should be treated accordingly. 17  This presumptive rule is sometimes called the “presumption of veracity” (e.g. Kauffeld 2003) and the “presumption of natural trustfulness” (Perelman & Olbrechts-Tyteca 1969, pp. 70-71).

29

CHAPTER 1

reasons.18 As the previous reconstruction shows, Smith has (sufficient) reasons for the presumption, even though he is not obliged to provide them.19 I have briefly presented one standard account of the structure underlying the foundations of presumption. But what justifies presumptions in qualitative terms? According to various legally-inspired authors, presumptions can be based on the status quo, as well as “fairness,” “eminent authority,” “persistent good judgment,” “the collective agreement,” “appearance of truth,” “suppositions about what is normal and likely,” “plausibility,” “the normative acceptability of error,” etc. (Kauffeld 2003, p. 138). The list becomes even longer if we take non-technical presumptions into account. In Kauffeld’s view, for instance, ordinary speakers infer presumption p from “the grounds that someone will have made that the case rather than risk criticism, painful regret, reprobation, lose of esteem [sic] or even punishment for failing to do so” (p. 140). However, as far as initial dialectical presumptions are concerned, it is crucial to understand how things normally work (or should work). Initial presumptions are supposed to be based on commonsensical, widely-accepted, and uncontroversial presumptive rules, such as: “Normally, we should trust our perception and memory,” “Normally, we should draw conclusions from experience,” “Normally, we should trust science and experts,” “Normally, we should form our expectations based on social conventions,” etc. Precisely because these rules are (or should be) accepted by reasonable interlocutors (and the basic facts are meant to be transparent), the opponent should be able to reconstruct the foundations of the initial presumption, and the proponent should not be (initially) obliged to provide reasons. Importantly, the inferential reconstruction of the foundations of presumption does not entail that all initial presumptions are, in fact, ‘inferentially justified.’ In contemporary epistemology, ‘inferential justification’ is a technical term that establishes the connection between the justification of a belief and doxastic states—a belief p is inferentially justified if and only if it logically follows (either deductively or inductively) from some set of justified beliefs (see Feldman 2003, p. 50; Hasan & Fumerton 2016). By contrast, on Godden & Walton’s (2007, p. 314) inferential account, presumptions can also be justified by nondoxastic reasons such as perceptual states. For instance, one’s contention “Presumably, 18 

According to the distinction offered by Aijaz, McKeown-Green & Webster (2013), an initial presumption may

unevenly allocate the ‘dialectical burden,’ i.e. exempt the proponent from providing (sufficient) argumentative reasons, but the proponent may still need to satisfy an ‘attitudinal burden’ of having (sufficient) reasons. 19  Of course, foundations of initial presumptions may (once again) surface during the dialogue. First, in some cases, the opponent’s critical reaction may reveal that she made an error in reconstructing the presumption’s foundations. Second, the opponent may explicitly ask the proponent to expound the foundations of presumption. In both cases, the proponent may provide explicit reasons for presumption, but this does not entail that she accepts the burden of proof. Proponents may offer argumentative reasons because they “find it practically useful to substantiate their position even though they may be under no particular obligation to do so” (Kauffeld 1998, p. 263). Alternatively, proponents may offer reasons that are not argumentative in the strict sense. For instance, the proponent would not offer reasons to argue that the presumption is acceptable, i.e., to persuade the opponent, but to explain why it is acceptable. Its acceptability can well be explained, but it still does not need the support of arguments.

30

Presumptions, and How They Relate to Arguments from Ignorance

there are clouds in the sky” is usually not justified due to some other justified belief, but due to the perceptual state of seeing clouds in the sky. Hence, Godden & Walton’s inferential reconstruction, which I rely on, is mainly a theoretical tool used to analyse and express the foundations of presumptions, and it is compatible with the view that some initial presumptions are not, in fact, ‘inferentially justified,’ i.e., that they do not arise inferentially from doxastic states. Initial presumptions are based on prima facie foundations20 that may include both doxastic and non-doxastic states (reasons).

1

2.3 Dialectical Rights and Obligations in Presumption Criticism My earlier characterization of presumptive reasoning may shed some light on the dialogical rights involved in criticism of presumption. The opponent’s obligation to criticize the acceptability of a presumption includes at least three particular rights—she is allowed to (1) attack the presumption (conclusion) directly, (2) attack the basic fact (premise), and (3) attack the application of the presumptive rule (conditional).21 Relying on Pollock’s terminology concerning epistemic defeaters (e.g., Pollock 1987), I will illustrate these strategies on our introductory example. First, Jones can try attacking the presumption directly and thereby ‘rebut’ presumptive reasoning. This strategy requires an argument against the presumed proposition that shows “The shortest way to reach the city center…” to be false. Jones may argue that the second street to the right is not the shortest way to reach the city center, since it leads them outside the city. Second, Jones can attack the basic fact. Pollock does not discuss this strategy, but other argumentation scholars usually call it the “premise tenability criticism” (van Laar & Krabbe 2013, p. 204). Jones may argue that Jill misunderstood Smith’s question. As a result, she did not testify “The shortest way to reach the city center is...”, but “The shortest way to reach the City Center is...”, where ‘City Center’ is the name of a shopping mall. Finally, Jones may try to ‘undercut’ the reasoning by showing that the application of the defeasible presumptive rule is not reliable in the particular case. For instance, if Jones suddenly recalls 20 

In this chapter, a prima facie foundation (or prima facie reason) for p is taken to be any kind of foundation (reason)

for p that is susceptible to defeaters, i.e., a reason that justifies p only in the absence of defeaters. Therefore, prima facie justification is not “justification 'on the first look' but rather justification 'other things being equal'” (Senor 1996, p. 554). 21  This set of attacking (defeating) strategies is not exhaustive. In our example, the presumptive rule is “Normally, we should trust other people.” The opponent may accept this rule and attack its application in a particular case. Following Pollock, I call this the undercutting strategy. The opponent is also allowed, at least in principle, to attack the tenability of the presumptive rule, e.g., to argue that we should not normally trust other people. Although the tenability criticism of the presumptive rule is legitimate, I do not include it in the analysis for two reasons. First, presumptive rules tend to be so basic that, in everyday contexts, where ordinary presumptions are made, their tenability is rarely under attack. For instance, the general tenability of “Normally, we should trust our senses” is hardly questioned anywhere but in epistemological discussions. Second, the arguments in this chapter do not depend on the analysis of this strategy. Thus, I will leave it aside for the sake of simplicity.

31

CHAPTER 1

that the locals often deceive tourists for fun, then Smith cannot legitimately conclude that the directions are correct solely on the basis of Jill telling him so. His reasoning is not rebutted, for the falsity of the presumption has not been indicated: Jill’s directions can be both correct and accepted on unreliable grounds. The presumption is, however, undercut, for Smith now lacks reliable foundations indicating that Jill’s testimony is true. Importantly, the undercutting strategy is not identical to the tenability criticism of a presumptive rule. The presumptive rule, stating that “Normally, we should trust other people,” is not at all under attack. Both parties still agree that we should usually trust other people. It is just that new information (the unusual habit of locals to deceive tourists) prevents the application of the presumptive rule. If the strong similarity thesis were true, then the aforementioned dialectical rights and obligations would be expected in dialogical settings where arguments from ignorance are typically used. In Section 4.4, I will argue that this expectation is ultimately unwarranted, which makes the strong similarity thesis unconvincing. But to get to this point, I first provide a brief elaboration of the argument from ignorance and discuss some supradialogical differences between initial presumptions arguments from ignorance.

3 The Argument from Ignorance: The Standard View Contemporary literature recognizes at least two types of argument from ignorance. The original type introduced by John Locke is only briefly explained in his Essay Concerning Human Understanding. In Locke’s words: Another way that men ordinarily use to drive others, and force them to submit their judgments, and receive the opinion in debate, is to require the adversary to admit what they allege as a proof, or to assign a better. And this I call argumentum ad ignorantiam. (Locke 1995, p. 56) However, my analysis in this chapter starts from the more common type of argument from ignorance normally found in logical textbooks.22 The so-called textbook argument from ignorance (TAI) has a somewhat different form: (TAI)

It is not known/proven that p (non-p)



Therefore, non-p (p).

[ignorance premise]

TAI has been traditionally known as an informal fallacy. For instance, the absence of evidence for “The bar sells orange juice,” cannot, in principle, serve as a good reason for the conclusion “The bar does not sell orange juice.”23 Scholars have often expressed this point 22 

For a detailed analysis of both types of argument from ignorance, see Krabbe (1995).

23 

For instance, if Smith wasn’t even searching for evidence that the particular bar sells orange juice, i.e., if he

hasn’t checked the bar’s price list, then the absence of evidence for “The bar sells orange juice” would hardly justify his conclusion “The bar does not sell orange juice.”

32

Presumptions, and How They Relate to Arguments from Ignorance

by stating that the absence of evidence is not evidence of absence. Although this is generally true, TAI can also be a reasonable argument in both epistemic and practical contexts.24 Let us start from an ignorance premise “It is not known that the water is contaminated” to conclude “The water is not contaminated.” If experts have carefully examined the quality of the water, then the following conditional is acceptable: “If the water were contaminated, then the contamination would have been found.” Assuming that experts have examined the quality of the water and did not find any contamination, it is plausible to conclude that the water is not contaminated. In terms of structure, we would conclude that (1) “A is not true” from the ignorance premise (2) “A is not known to be true” and a conditional (3) “If A were true, then A would be known to be true” (Walton et al. 2008, p. 98).25 Crucially, without premise (4) “Experts have carefully examined the quality of the water,” the previous conditional would hardly be acceptable—the connection between the ignorance premise and the conclusion would not be reliable. With this premise backing the conditional, however, our conclusion will be derived from a good “epistemic argument from ignorance” (Walton et al. 2008, p. 99) or a good argument based on “negative evidence” (Macagno & Walton 2011, p. 95).

1

TAI can also be a good practical argument. Suppose that the outcome of deliberation depends on p, yet p is evidentially uncertain. If proceeding on p avoids serious consequences, the absence of evidence for non-p is a plausible reason to conclude p. In Walton’s wellknown example, we need to decide how to handle a weapon, yet it is uncertain whether the weapon is loaded. Making the wrong decision can be very costly, but we are aware that proceeding on “The weapon is loaded” reduces the risk of serious (and potentially fatal) consequences. Hence, until we are completely certain that the weapon is not loaded, we will conclude that it is loaded. Here, the argument from ignorance is an instance of “negative practical reasoning” (Walton et al. 2008, pp. 99-100). The contaminated water example and the loaded weapon example indicate that there is nothing intrinsically wrong with arguments from ignorance. As Hahn & Oaksford (2007, p. 53) remark, there will be “cases when they are weak and cases when they are strong, as with other types of argument.” This outcome coheres rather well with the strong similarity thesis, since initial dialectical presumptions are never described as fallacious in the literature. Before proceeding further, it should be stressed that the practical version of TAI is indeed similar to (paradigmatic) practical presumptions. According to Ullmann-Margalit (1983), Godden (2017) and, occasionally, Walton (2008b, 2014), presumptions are nonepistemic means used in deliberation. The proponent is not obliged to prove p (while the 24  For a discussion on the epistemic and practical versions of TAI, see, e.g., Wreen (1989), Krabbe (1995), Walton, Reed & Macagno (2008), and Tuzet (2015). 25 

According to this counterfactual formulation of the conditional (connection premise), the ignorance premise is a

negation of the consequent. Hence, the conclusion follows from (defeasible) modus tollens. However, it is also intuitive to read the ignorance premise as the antecedent of the conditional “If A is not known to be true, then A is not true.” In this case, the conclusion follows from (defeasible) modus ponens.

33

CHAPTER 1

opponent is obliged to prove non-p) when there is pressure to make a decision, this decision depends on whether p is true, p is uncertain but proceeding on p is (more) practically justified (than proceeding on non-p). Since a decision is surely (more) practically justified if it avoids serious consequences, the loaded weapon example is paradigmatic of practical presumptive reasoning. In effect, one version of the strong similarity thesis appears to be true—(a paradigmatic type of) practical presumptive reasoning is identical to the (practical) argument from ignorance. However, this chapter is not concerned with practical presumptions and practical arguments from ignorance. It analyses the similarity between initial epistemic presumptions and epistemic arguments from ignorance. In other words, it explores whether the similarity that appears to hold for ‘practical cases’ also holds for ‘epistemic cases’ of the two phenomena.

4 Is the Strong Similarity Thesis Tenable? In this section, I explore the relationship between presumptions and arguments from ignorance by testing the strong similarity thesis. I will argue against strong similarity— typically, presumptions and arguments from ignorance are different from both a structural and a normative perspective, since they involve different foundations (Section 4.2), and different defeasibility conditions (Section 4.3). Since they bring about different dialectical rights and obligations, they are different from a dialogical perspective as well (Section 4.4). As I already mentioned, I construct the strong similarity thesis (SST) as a conjunction of two entailments—the ignorance-presumption entailment (IPE) and the presumptionignorance entailment (PIE). Before putting strong similarity to this test, however, the meaning of ‘similarity’ needs to be clarified. In what respects are initial presumptions and textbook arguments from ignorance (TAI) supposed to be similar? In Section 4.1, I propose four possible answers to this question. Once the most plausible interpretation of similarity is selected, I explore in some detail whether this particular interpretation renders initial presumptions and TAI strongly similar.

4.1 The Interpretations of Similarity To make the notion of similarity more explicit, I begin by focusing on the absence of positive reasons. Consider the following interpretation: (S1) There are no positive reasons supporting either the presumption or the conclusion of TAI. Yet, they are both (allegedly) acceptable. S1 is the least plausible interpretation of similarity. As I explained in Section 2.2, the acceptability of an initial presumption depends on prima facie foundations that are typically constituted by positive reasons. These reasons play a crucial role for the epistemic acceptability of p, and are also normatively relevant for the acceptance of a dialogical

34

Presumptions, and How They Relate to Arguments from Ignorance

presumption.26 In effect, S1 is false—the initial presumption cannot be similar to TAI due to the lack of positive reasons to support it. Since there is no similarity, there cannot be strong similarity either. However, there is a simple solution to the latter problem. We only need to shift our focus from the mere existence of positive reasons to their explicit presence in the dialogue. Consider the following: (S2) The proponent does not provide any positive reason supporting either the presumption or the conclusion of TAI. Yet, they are both (allegedly) acceptable.

1

S2 offers some improvement but is, nevertheless, false. As the contaminated water example shows, epistemic arguments from ignorance lean on positive reasons quite explicitly and it is not clear why the proponent could not provide the positive fact (i.e., that experts have examined the quality of the water) as a reason to support the conditional premise. Admittedly, this positive reason is not an immediate (direct) reason for the conclusion, but it is still necessary for its acceptability. Hence, S2 is not a plausible interpretation, and since there is no similarity, there cannot be strong similarity either. Although it renders S2 implausible, the contaminated water example provides clear guidelines for constructing the third interpretation of similarity. (S3) The proponent does not provide and is not obliged to provide any immediate positive reason supporting either the presumption or the conclusion of TAI. Yet, they are both (allegedly) acceptable. This interpretation of similarity is true, but it is also misleading. By stressing that the proponent does not provide, and is not obliged to provide an immediate positive reason, S3 obscures the fact that the proponent typically does not provide (and she is, by definition, not obliged to provide) any argumentative reason (either positive or negative) in support of the initial presumption. By contrast, when arguing from ignorance, the proponent typically does provide a negative reason to support her conclusion and to meet the burden of proof. This difference reveals a crucial difference between dialectical obligations—whereas a conclusion of the argument from ignorance typically does not exempt the proponent from the burden of proof, the initial presumption is, by definition, dialectically privileged.27 So, 26  Of course, one may insist that some (sufficient positive) reason would be normatively relevant only if the proponent were obliged to provide this reason in the dialogue. But this is obviously problematic. For instance, if the conclusion “You should stop smoking” (P) hinges on the tacit premise that smoking is not healthy, and this is recognized by all parties in the dialogue, then the proponent may not be obliged to provide “Smoking is not healthy” (Q) as an explicit argumentative reason. However, this does not render Q irrelevant for the dialogical acceptability of P. If this point works for ordinary enthymemic argumentation, why should it not apply in the case of presumption? 27 

As I mentioned in the introduction, some scholars suggest that (fallacious) arguments from ignorance (illicitly)

shift the burden of proof. I do not think this is true. In Sect. 4.4, while relying on Krabbe’s (1995) terminology, I briefly explain the difference between an ‘appeal to ignorance’ that shifts the burden of proof (without providing a reason) and the textbook argument from ignorance (where the proponent provides a negative reason).

35

CHAPTER 1

as long as conditions and obligations regarding the provision of reasons are concerned, S3 seems to select only those aspects that are in the best interest of similarity. Since this is rather problematic, S3 can hardly help us advance a clear-headed study of presumptions, and is, thereby, unsuitable for our purposes. The most promising interpretation of similarity concerns not the (dialogical) irrelevance of positive reasons, but the normative relevance of defeaters. We can phrase this as follows: (S4) Both the acceptability of presumption p, and the acceptability of TAI’s conclusion p ‘depend on’ the absence of a ‘sufficiently strong defeater’ (offered by the opponent). This interpretation of similarity is intuitive, accurate, and unbiased, which makes it a suitable starting point for testing the strong similarity thesis (SST). However, in the following sections, I will argue that the vagueness of phrases ‘depends on’ and ‘sufficiently strong defeater’ weakens the proposal, i.e., that once these notions are elucidated, the initial epistemic presumption and the epistemic argument from ignorance part ways and become different from structural, normative, and dialogical perspectives. In other words, I will argue that although S4 is true, the SST is false.

4.2 The Structural Perspective: The Nature of Dependency Relations To understand why the notion of ‘dependency’ is vague, suppose that p is the conclusion of a (defeasible) modus ponens, and c is any condition that contributes to p’s acceptability. Defined in this way, c can represent different elements in the structure of the argument. In the simplest case, c can be either the conditional or its antecedent. In a more complex case, c can be some reason that supports the conditional or, perhaps, the reason that supports the antecedent. Accordingly, the notion of ‘dependency’ is ambiguous and the statement “p depends on c” is vague: the conclusion p can depend on c either directly (c is the conditional or an antecedent) or indirectly (c is a reason that supports the conditional or a reason that supports an antecedent). This ambiguity explains why presumptive reasoning and the argument from ignorance (TAI) are not strongly similar from the structural perspective. Suppose that the acceptability condition C, relevant for our present purposes, is defined as ‘the absence of a sufficiently strong defeater.’ In our introductory example of an initial presumption, C represents neither the basic fact (antecedent) nor the presumptive rule (conditional). Instead, it specifies the circumstances in which it is reasonable for Smith to presume “The shortest way to reach the city center...” given Jill’s testimony.28 Thus, C is only indirectly relevant for the acceptability of Smith’s presumption. Call this the ‘typical presumptiondependency relation’ (PD), where the expression ‘depends on’ is conveniently substituted. (PD) The acceptability of (a typical presumption) p [indirectly depends on] C. 28 

Instead of a basic fact or a presumptive rule, C appears to be a necessary condition for the reasonable application of

a presumptive rule.

36

Presumptions, and How They Relate to Arguments from Ignorance

By contrast, the acceptability of conclusion p of the argument from ignorance (TAI) directly depends on C, at least in part.29 As we have seen, the so-called ‘ignorance premise’ in TAI notes the absence of a sufficiently strong rebutting defeater. Since the ignorance premise is a conditional’s antecedent (or, in the alternative formulation, a negation of the conditional’s consequent), it follows that C is directly relevant for the acceptability of p. Call this the ‘TAI dependency relation’ (TD): (TD) The acceptability of p [directly depends on] C.

1

Both Bermejo-Luque and Rescher highlight the difference between the two dependency relations. Bermejo-Luque observes that “presumptive inferences appeal to ignorance in order to point out that a possible defeater for the warrant (which is a presumption) does not obtain, whereas typical arguments from ignorance characteristically appeal to ignorance as the very reason to conclude” (2016, p. 21). Rescher makes the same point, noting that in presumptive reasoning, “ignorance is not a ground or premise for which to reason but a circumstance in which one reasons as best one can” (2006, p. 2). Rescher concludes that the presumptive argument is an argument in ignorance rather than an argument from ignorance. Thus, typical initial presumptions and textbook ad ignorantiam ‘depend’ on the absence of defeaters in structurally different ways. However, initial dialectical presumptions are not always typical. Some initial presumptions may well be conclusions of textbook ad ignorantiam. Acknowledging the existence of these exceptional cases will lead to a more inclusive account of the dependency relation. Call this the ‘complete presumption-dependency relation’ (CPD): (CPD) The acceptability of (presumption) p [either directly or indirectly depends on] C. CPD may grant the strong similarity thesis (SST) another chance. However, the comparison between CPD and TD renders SST untenable. To understand this, we must recall that the tenability of SST requires the tenability of both the ignorance-presumption entailment (IPE) and the presumption-ignorance entailment (PIE). Consider IPE first. If the textbook ad ignorantiam entails presumptive reasoning from the perspective of dependency relations, then the dependency relation expressed by TD should entail the dependency relation expressed by CPD. This happens to be the case. Namely, if p directly depends on C, then we may conclude “Either p directly depends on C, or p indirectly depends on C.” Nevertheless, if the strong similarity thesis holds from the perspective of dependency relations, then the dependency relation in CPD should entail the dependency relation in TD. But this entailment does not hold. It does not follow from “Either p directly depends on C, or p indirectly depends on C” that p directly depends on C. Since we assumed that both PIE and IPE are necessary to claim strong 29 

The acceptability of TAI’s conclusion p directly depends on the absence of a rebutting defeater. By contrast, the

absence of an undercutting defeater appears to be a necessary condition for the reliability of a connection premise (conditional). As a result, the acceptability of p also indirectly depends on the absence of an undercutting defeater.

37

CHAPTER 1

similarity, SST does not hold when the dependency relations of presumption and textbook ad ignorantiam are concerned. My analysis so far worked under the assumption that presumption and argument from ignorance depend on the same acceptability condition C, and explained why they do not depend on C in a structurally similar way. Simply put, while the argument from ignorance is, by definition, a type of negative inference, typical presumptive reasoning is not. In the next section, I deepen the analysis of similarity by scrutinizing the acceptability condition C. I argue that the notion of a ‘sufficiently strong defeater’ is also vague and that, as a result, the strong similarity thesis (SST) is not plausible from a normative perspective.

4.3 The Normative Perspective: The Content of Acceptability Conditions I have proposed earlier that presumptions and arguments from ignorance are similar because their acceptability depends on ‘the absence of a sufficiently strong defeater,’ i.e., on C. Call this the ‘similar acceptability condition’ (SA). (SA) The acceptability of p depends on [the absence of a sufficiently strong defeater].30 This statement is uncontroversial. Regardless of how we interpret p in SA (as a presumption or as a conclusion of the argument from ignorance), SA seems to hold, and a similarity exists. Nevertheless, the dependency on the absence of defeaters is surely not sufficient to render presumptive reasoning and the argument from ignorance similar in an intriguing way. After all, in any non-deductive (or non-monotonic, defeasible) argument, the acceptability of the conclusion depends on the absence of defeaters, but this hardly makes different kinds of non-deductive arguments strongly similar. For instance, it would be counterintuitive to claim that enumerative induction is strongly similar (let alone identical) to statistical syllogism, or that it is difficult to distinguish an argument from expert opinion from an argument from analogy. This aspect of similarity is too general to make the strong similarity thesis theoretically interesting. More importantly, once we unpack the vague notion of ‘a sufficiently strong defeater,’ it becomes clear that presumption and argument from ignorance appeal to different sets of defeaters. Whereas a typical initial presumption is susceptible to three types of defeaters (rebutting defeater, undercutting defeater, premise tenability criticism), the conclusion of an ad ignorantiam depends only on the absence of a rebutting defeater and an undercutting defeater.31 In other words, the ‘typical presumption acceptability 30 

Notice that SA is but a mere rephrasing of S4.

31  Due to the content of an ignorance premise, one may come to believe that the argument from ignorance is susceptible only to the rebutting defeater. This is, however, false. As the contaminated water example has shown, some arguments from ignorance explicitly rest on indications that the connection between the ignorance premise and the conclusion is reliable—the fact that experts examined the water increases the reliability of the connection between “It is not known

38

Presumptions, and How They Relate to Arguments from Ignorance

condition’ (PA) and the ‘ignorance acceptability condition’ (IA) are different instantiations of the ‘similar acceptability condition’ (SA). (PA) The acceptability of (the typical presumption) p depends on [p is not rebutted, q is not proven untenable, and the connection between q and p is not undercut]. (IA) The acceptability of the conclusion p depends on [p is not rebutted, and the connection between q and p is not undercut].

1

As soon as a less abstract comparison is made, presumption and the argument from ignorance part ways—they become easily distinguishable, and the strong similarity thesis (SST) ceases to hold. Let’s explain this verdict in terms of the ignorance-presumption entailment (IPE). If the acceptability conditions assumed by the argument from ignorance were strongly similar to the ones assumed by presumptive reasoning, then satisfying the acceptability condition in IA would entail satisfying the acceptability condition in PA. But this is hardly true—there is no reason to think that “p is not rebutted and the connection between q and p is not undercut” entails “q is not proven untenable.”32 Since IPE is untenable and both PIE and IPE are necessary to ensure strong similarity, SST does not hold with regard to the acceptability conditions of presumption and ad ignorantiam. However, there are at least two strategies that the proponent of SST can use in response to my previous argument. First, she could argue that, in principle, IA is not complete (correct). This strategy would restore strong similarity by showing that, in the argument from ignorance, all three types of a sufficiently strong defeater are recognized. The advocate of SST may argue that the core of the argument from ignorance is the ignorance premise, and that the conclusion will be acceptable only if the ignorance premise is tenable. Thus, the complete version of IA would include the same set of defeaters as PA and, as a result, the strong similarity thesis would hold. The problem with this strategy is that, in the case of the argument from ignorance, we cannot easily distinguish the premise tenability criticism from the rebutting defeater. To see this, consider what the tenability criticism of an ignorance premise “There is no sufficient reason for non-p” may require. To render this ignorance premise untenable, I believe, we must offer at least one reason for non-p. But offering such a reason comes down to an attempt to prove the contrary, and the latter is the very definition of a rebutting defeater. Therefore, in the particular case of ad ignorantiam, the tenability criticism of the ignorance premise is typically reducible to the existence of a rebutting defeater. In effect, that the water is contaminated” and the conclusion “The water is not contaminated.” Given that, in these arguments, the normative relevance of conditions increasing the connection’s reliability is recognized, the normative significance of conditions decreasing the connection’s reliability should also be given recognition. The undercutting defeater is precisely the condition of the latter kind. 32 

By contrast, PIE is plausible—satisfying the acceptability condition in PA entails satisfying the acceptability

condition in IA. Namely, once the acceptability condition for the presumption is satisfied, presumption p is neither rebutted nor undercut, which means that the acceptability condition for the ad ignorantiam is satisfied too.

39

CHAPTER 1

the acceptability conditions of the two remain easily distinguishable: unlike typical initial presumption, the conclusion of the argument from ignorance cannot depend on three types of defeaters in any genuine or informative sense.33 However, the advocate of SST has another strategy at her disposal. She may try restoring the similarity by claiming that, sometimes, PA is not correct. Since some initial presumptions are conclusions of the epistemic arguments from ignorance, some initial presumptions are defeated only by rebutting and undercutting. Consequently, PA and IA (sometimes) include the same acceptability condition. Although this is true, this is a rather desperate attempt at restoring strong similarity. Exceptional cases cannot support the strong similarity thesis—that some initial presumptions are epistemic arguments from ignorance does not imply that this is typically the case. Some proposition may (still) have the acceptability condition of a (typical) presumption, but not the acceptability condition of an argument from ignorance. Hence, since PIE is untenable, SST is untenable as well. To sum up, not only do typical initial presumptions and textbook arguments from ignorance include structurally different foundations, but they also have different acceptability conditions. As far as the normative role of defeaters is concerned, the conclusion of a typical ad ignorantiam seems to be privileged compared to a typical initial presumption: whereas the absence of both rebutting and undercutting defeaters is sufficient to make the conclusion of an ad ignorantiam acceptable, it is only necessary (but, typically, not sufficient) for presumption to be acceptable. In some sense, typical initial presumptions have the higher negative standard of acceptability—there is a broader set of constraints, a broader set of genuine restrictions that the proponent will (possibly) need to accommodate. There is a lesson to be learned here concerning the nature and limits of presumption’s privileged status. Intuitively, we can render p normatively privileged either by making it easy to justify p or by making it hard to defeat p. It seems that a typical initial presumption is privileged in the first sense but not in the second. In dialogue, the proponent is not obliged to provide sufficient reasons to render a presumption acceptable, which makes it, in a dialectical sense, ‘easy to justify.’ However, the initial presumption is also easily defeated.34 The 33 

I do not claim that the tenability criticism and the rebutting defeater are coextensive. One could, perhaps, offer a

rebutting defeater without offering a tenability criticism of a particular ignorance premise. For instance, I may believe that the water is contaminated because I have drunk some of it earlier and now I feel sick. This may, perhaps, rebut the conclusion “Presumably, the water is not contaminated” without showing that the ignorance premise “Our scientific methods did not detect the contamination” is untenable. Rather, my point is that the opposite is hardly possible—one cannot offer evidence that we detected contamination (i.e., tenability criticism) without offering some evidence that water is contaminated (rebuttal). Importantly, this is not to say that good tenability criticism is always a good rebutting defeater— some counterexamples are possible here. However, the opponent who is trying to prove that some contamination was detected is surely ‘in the same line of business’ as someone who is trying to prove, either successfully or unsuccessfully, that the water is contaminated. I thank the reviewer for pressing this clarification. 34  By saying that it is easy to defeat a typical presumption, I only mean that different defeating strategies are available. This (formal) view is compatible with the fact that, in reality, defeating presumptions may indeed be difficult, since

40

Presumptions, and How They Relate to Arguments from Ignorance

opponent has (at least) three available strategies to defeat the typical initial presumption and (arguably) only two available strategies to defeat the conclusion of epistemic ad ignorantiam. This shows that initial presumptions are not so privileged when defeating conditions are concerned. It is quite common to characterise presumption in terms of its privileged status, and this analysis clarifies the nature and limits of this privilege. In the final section, I test SST by comparing attacking rights and obligations that arguments from ignorance and presumptive reasoning impose in dialogical contexts. Since most theories of argumentation regard dialectical rights and obligations as essential in defining presumptions and presumptive reasoning (see Walton 2014, 2008b; Godden & Walton 2007; Rescher 2006, 1977; Freeman 2005; Pinto 2001, etc.), the dialogical explanation is indeed theoretically relevant. Here, the dialogical analysis represents the implementation or extension of the normative perspective (i.e., the perspective of acceptability conditions).

1

4.4 SST and Dialectical Rights and Obligations In his analysis of appeals to ignorance, Krabbe (1995) suggests that the proponent may respond to the opponent’s challenge in (at least) two characteristic ways: by introducing a counterchallenge, or by asserting an ignorance premise. This means that features of appeals to ignorance can be studied in (at least) two characteristic dialogue fragments. In the first fragment, the proponent responds by introducing a counterchallenge. P:

p

O:

Why p?

P:

Why non-p?

By introducing the counterchallenge “Why non-p?” the proponent shifts the burden of proof. She refuses to offer a reason for p and demands an argument that would prove otherwise. This is one way of saying: “My contention p stands until you show otherwise.”35 Accordingly, this dialogue fragment assigns rights and obligations to the opponent in the following way. In the next move, O is allowed both to concede p and to attack p (by rebutting it). O is also obliged (at least from the proponent’s perspective) to choose between available options: she must either concede p or attack p (by rebutting it). If O refuses to concede p, then she becomes obliged to rebut p, i.e., she must provide a sufficiently strong argument for non-p. presumptions are usually based on widely-accepted commonsensical rules and transparent facts, and the standard of proof is set very high. 35  This particular kind of appeal to ignorance is usually portrayed as a dialectical fallacy, since the proponent responds to a legitimate challenge by illicitly shifting the burden of proof (e.g., Woods & Walton 1978). In addition, Krabbe argues that the proponent is committing a straw man fallacy. Namely, the opponent is asked to prove something she did not claim: to prove non-p despite only expressing doubt regarding p. Hence, this kind of appeal to ignorance involves a shift of attitude or an internal dialectical shift (Krabbe 1995, p. 256).

41

CHAPTER 1

The first dialogue fragment may raise some concerns regarding my previous characterization of the argument from ignorance (TAI). Notice that, in the first fragment, the proponent does not provide any reason for the acceptance of p. By employing a counterchallenge, she requests reasons, which is quite different from providing them. So, one may come to believe that my structural and normative characterizations of TAI were incorrect. From the structural perspective, I proposed that p directly depends on an ignorance premise, but this cannot be true if there is no premise. From the normative perspective, I proposed that p is susceptible to an undercutting defeater, but this cannot be true if there is no connection to be attacked.36 However, I have one simple reason to think that these concerns do not threaten the previous characterization of TAI: this fragment expresses an appeal to ignorance, but it does not lay out an argument from ignorance. Since the proponent does not offer any reason for a conclusion, she does not offer an argument, and since there is no argument—there is no argument from ignorance either. Accordingly, this fragment should affect neither my characterization of TAI nor any previous conclusions regarding the similarity between TAI and presumptive reasoning.37 In another case of appealing to ignorance, the proponent responds to the opponent’s challenge by offering an explicit ignorance premise. The following fragment expresses the dialogical implementation of TAI. P:

p

O:

Why p?

P:

You don’t have a sufficient reason for non-p.

Once the distribution of the opponent’s dialectical rights and obligations in the next, fourth step of the dialogue is concerned, this dialogue fragment is different from the previous one: O is now also allowed to undercut P’s argument. Let us examine the fragment in more detail. Once again, O is allowed to concede p and to attack p by rebutting it. O is also not allowed to attack P’s reason for p in any genuine or informative sense. That is, although P offers an argument based upon an explicit ignorance premise, criticism of the ignorance premise may be reduced to the rebutting strategy (as explained in Section 4.3). However, O is allowed to undercut P’s argument—in the subsequent steps of the dialogue, O can legitimately contend that her lack of sufficient reasons for non-p is not a reliable indicator that p is acceptable, and she can support this contention by offering some concrete reason r. 36 

Clearly, if there is no premise (reason), then there is no connection between a premise and a conclusion, and if there

is no connection, then there is no undercutting defeater. 37 

To give this fragment its due, perhaps we should treat it as a dialogical precondition, or a dialogical root of TAI.

Suppose that in the fourth step, immediately following the proponent’s counterchallenge, the opponent remains silent—she simply does not respond to the proponent’s counterchallenge. In the fifth step, the proponent makes her ex silentio move: “Since you have not proven non-p, p is acceptable.” In this extended version of the fragment, however, the proponent draws her conclusion from the ignorance premise. This means that my characterization of TAI is accurate and, once again, there is no reason to doubt my findings concerning the similarity between TAI and typical presumptive reasoning.

42

Presumptions, and How They Relate to Arguments from Ignorance

Finally, O is obliged to choose between available options: she must either concede p or attack p. If O refuses to concede p, then she becomes obliged to offer a sufficient argument against the acceptance of p. The latter argument can contain a rebutting or an undercutting defeater. It seems that the two kinds of appeals to ignorance entail different distributions of the opponent’s dialectical rights and obligations in the next, fourth step of the dialogue. However, each of these distributions is still different in a relevant way from the one produced by typical presumptive reasoning. I will explain the difference between typical presumptive reasoning and TAI in terms of the ignorance-presumption entailment (IPE) and the presumption-ignorance entailment (PIE).

1

Typical presumptive reasoning and TAI are different both in terms of O’s attacking rights and O’s attacking obligations. In typical presumptive reasoning, O is allowed to (1) rebut presumption p, (2) attack the tenability of premise q, and to (3) undercut the connection between q and p. In TAI, O is allowed to (1) rebut conclusion p and to (3) undercut the connection between an ignorance premise q and conclusion p. This indicates that PIE does, while IPE does not hold from the perspective of dialectical rights. In other words, it is possible to derive O’s rights to rebut the conclusion and to undercut the connection between the ignorance premise and the conclusion from the conjunction of O’s typical rights involved in presumption criticism. However, it is not possible to derive the larger conjunction of O’s typical rights involved in presumption criticism from only two members of this conjunction. The dialectical obligations are also different. Once O decides to attack presumption p, she is obliged to “Either (1) rebut p, or (2) attack the tenability of q, or (3) undercut the connection between q and p.” Once O decides to attack the ad ignorantiam, she is genuinely obliged to “Either (1) rebut p, or (3) undercut the connection between q and p.” This indicates that IPE does, while PIE does not hold from the perspective of dialectical obligations—it is not possible to derive “O is obliged to rebut p, or to undercut the connection between q and p” from the broader disjunction of O’s dialectical obligations (involved in presumption criticism). Since we assumed that both PIE and IPE are necessary to claim strong similarity, SST does not hold when rights and obligations related to presumptive reasoning and textbook ad ignorantiam are concerned. This is a dialogical way of saying that the argument from ignorance and typical presumptive reasoning include different acceptability conditions. Presumptions, as portrayed in this chapter, are essentially dialectical phenomena, so it is appropriate to study their features in dialogical settings. However, the relevance of the dialogical analysis does not lie only in its appropriateness, but it also enables us to deepen our understanding of the difference between presumptions and arguments from ignorance. It reveals that the difference is not only found in the sets of available attacking strategies but that these different sets also imply different conditional rights and obligations. Yet, even against the background of this richer dialogical ontology, the strong similarity thesis still does not hold—IPE does not hold from the perspective of dialectical rights, while PIE does not hold from the perspective of dialectical obligations. As a result, the typical initial 43

CHAPTER 1

presumption and the epistemic argument from ignorance are different from structural, normative, and dialogical perspectives.

5 What is the Upshot of the Previous Analysis? What is the ultimate upshot of this analysis? What are the (broader) theoretical implications of its results? Clearly, the analysis shows that initial presumptions and arguments from ignorance merit distinct treatment. They are similar in fewer respects than it may appear at first, and the similarities they do presumably share either arise from ambiguities, i.e., they are based on a problematic (biased) selection of features, or are simply too broad to be theoretically interesting. It follows that the strong similarity thesis is false. There are many differences between typical initial presumptions and arguments from ignorance, apparent from structural, normative, and dialogical perspectives. Apart from being false, the strong similarity thesis is also methodologically unhelpful. In other words, argumentation scholars often emphasize that a good theory of argumentation must provide methods for the analysis and evaluation of arguments. Accepting strong similarity often entails lumping together items with distinct inferential structures (during argument analysis) and distinct normative requirements (during argument evaluation). The strong similarity thesis also conceals that initial presumptions and arguments from ignorance belong to different stages of an argumentative discussion. While initial presumptions are associated with the very foundations of dialogue and are accepted in the ‘opening stage,’ arguments from ignorance typically represent a particular kind of argumentative strategy and typically emerge in the ‘argumentative stage’ of dialogue. Thus, it is rather difficult to anticipate the methodological benefits of accepting the strong similarity thesis. The analysis offered in this chapter helps us to dismiss possible misconceptions, which I find helpful both cognitively (theoretically) and methodologically. The analysis also contributes to the study of presumptions’ fundamental aspects, such as their foundations and force (strength) (Godden & Walton 2007). For instance, it makes clear (albeit in an indirect way) that initial (shared/epistemic/cognitive) presumptions and practical presumptions are quite different phenomena: while paradigmatic practical presumptions (as defined by Ullmann-Margalit and Godden) are meant to be structurally identical to (practical) arguments from ignorance, typical initial presumptions are not structurally identical to (epistemic) arguments from ignorance. This conclusion is directly relevant to the ongoing investigations of the different kinds of presumptions and their corresponding foundations. Finally, our analysis may grant us a new perspective for studying the strength of presumption. Scholars usually illustrate the strength of presumption in terms of how difficult it is (for the opponent) to overturn it—the more demanding the standard of proof for non-p,

44

Presumptions, and How They Relate to Arguments from Ignorance

the stronger the presumption p.38 However, our analysis points out that the standard of proof is not the only important factor. The difficulty of defeating a presumption could also depend on the particular set of defeaters that the opponent has at her disposal. Suppose, for instance, that p is a typical initial presumption and r is a presumption based on an ignorance premise. The opponent will have fewer ‘genuine’ attacking strategies to defeat r and, in some sense, this will make it more challenging for her to defeat r. Thus, all else being equal, presumption r will be stronger than presumption p: when two presumptions presuppose the same standard of proof, the one with a more restricted set of defeaters might be considered stronger than the other. Thus, our analysis of acceptability conditions may provide guidance for a more comprehensive theory of presumption strength.

1

The primary aim of this chapter, however, was to advance a clear-headed study of presumptions. In order to construct a robust theory of presumption, the relationship between presumptions and arguments from ignorance needed to be sorted out. Until now, this relationship lacked a thorough analysis—some authors acknowledged (some) similarity, others expressed their reservations, but none of them, to my knowledge, offered an elaborate analysis. Thus, the final potential benefit of this analysis is ‘dialectical’ in nature—it sought to advance the debate and add to a serious and well-structured discussion.

6 Conclusion According to Douglas Walton, presumptions and arguments from ignorance are “so closely entangled that it is hard to distinguish one from the other” (2008a, p. 50). In this chapter, I argued that the Waltonian intuition is either limited (at best) or ill-founded (at worst), given that initial presumptions and epistemic arguments from ignorance do not share any significant similarities. I showed that typical presumptive reasoning and the argument from ignorance are different from structural and normative perspectives. From a structural perspective, they have different foundations: a typical initial presumption only indirectly depends on the absence of a defeater, whereas the conclusion of an argument from ignorance is directly drawn from the absence of a rebutting defeater. From a normative perspective, initial presumptions are typically susceptible to a broader set of defeaters. Since they can be rendered unacceptable by the broader set of defeaters, their acceptability entails a higher (negative) standard. I have also demonstrated this in the dialogical context, in terms of the opponent’s attacking rights and obligations. The analysis identifies the limitations of the theoretical proposal occasionally suggested by some authors, which makes it, I believe, both theoretically and methodologically useful. The analysis also (indirectly) contributes to the ongoing investigation of the different kinds 38 

In other words, if the opponent can defeat p only by providing conclusive reasons for non-p, then the presumption p

will be stronger than some presumption r that can be defeated by any reason for non-r (see, e.g., Ullmann-Margalit 1983, p. 152; Rescher 2006, p. 18).

45

CHAPTER 1

of presumptions, their foundations, their conditions of defeat, and the estimations of strength. Finally, it seeks to initiate an analysis of the theoretical proposal which, despite its presence in the literature, has not been carefully analysed so far. Acknowledgments: Some parts of this article were presented at the workshop Argumentations and conversations (March 13 2017, University of Granada, Spain) and at the PCCP meeting (April 5 2018, Faculty of Philosophy-University of Groningen, The Netherlands). I thank the organizers and the participants for their comments and constructive criticisms. I am especially indebted to Jan Albert van Laar who made many helpful suggestions on improving the earlier versions of the article. I am also grateful to Catarina Dutilh Novaes, Barteld Kooi, Karolina Kudlek, and Diego Castro Amenábar. Finally, I express my sincere gratitude to the editors and anonymous reviewers of Argumentation for their thorough analysis, detailed and constructive response, benevolence, and patience.

46

Presumptions, and How They Relate to Arguments from Ignorance

1

47

2

CHAPTER 2 On the Differences Between Practical and Cognitive Presumptions

CHAPTER 2

Words were not given to man in order to conceal his thoughts. —Saramago, The Cave

50

On the Differences Between Practical and Cognitive Presumptions

On the Differences Between Practical and Cognitive Presumptions Abstract: The study of presumptions has intensified in argumentation theory over the last years. Although scholars put forward different accounts, they mostly agree that presumptions can be studied in deliberative and epistemic contexts, have distinct contextual functions (guiding decisions vs. acquiring information), and promote different kinds of goals (non-epistemic vs. epistemic). Accordingly, there are ‘practical’ and ‘cognitive’ presumptions. In this chapter, I show that the differences between practical and cognitive presumptions go far beyond contextual considerations. The central aim is to explore Nicholas Rescher’s contention that both types of presumptions have a closely analogous pragmatic function, i.e., that practical and cognitive presumptions are made to avoid greater harm in circumstances of epistemic uncertainty. By comparing schemes of practical and cognitive reasoning, I show that Rescher’s contention requires qualifications. Moreover, not only do practical and cognitive presumptions have distinct pragmatic functions, but they also perform different dialogical functions (enabling progress vs. preventing regress) and, in some circumstances, cannot be defeated by the same kinds of evidence. Hence, I conclude that the two classes of presumptions merit distinct treatment in argumentation theory.

2

Keywords: avoiding harm, cognitive presumption, dialectical regress, Nicholas Rescher, practical presumption, presumptive reasoning, scepticism, uncertainty, undercutting defeater.

1 Introduction39 Suppose Anne and Jim are meeting a friend on a cloudy winter day. Just before leaving the apartment, they start deliberating whether to take an umbrella. They are aware that their decision should partly depend on whether it will rain—if rain is significantly more likely, they should take an umbrella; if rain is significantly less likely, they clearly should do the opposite. Anne quickly looks through the window and estimates that the chance of rain is, roughly, a half. She then checks a (fairly reliable) weather forecast website only to learn that the likelihood of rain is indeed 50%. Although Anne remains in a state of evidential uncertainty, the pressure of making a decision is forced upon her. She and Jim need to decide whether to take an umbrella, and they need to do it immediately (or risk being late). Next, suppose that Andy is an epistemically responsible agent who wants to know where his cat is. After looking through the window and seeing that a cat (which looks exactly like his cat) is in the tree, he expresses his belief that “Andy’s cat is in the tree” 39 

Paragraphs and sections of this chapter, which present typical features of practical and cognitive presumptions, are

partly based on Bodlović (2019a, 2020a). In this chapter, I develop a more detailed account of these features by elaborating the differences between practical and cognitive presumptions.

51

CHAPTER 2

thereby committing himself to its acceptability. However, Andy’s wife Jane is a radical sceptic. She challenges Andy’s commitment and, in good Cartesian fashion, remarks that Andy can never rule out the possibility of perceptual deception: from her sceptical viewpoint, the general uncertainty of perception renders the proposition “Andy’s cat is in the tree” evidentially inconclusive. This sceptical remark puts cognitive pressure on Andy. Not only must he reconsider whether his commitment is justified and whether it can be used to draw tentative conclusions (e.g., “Andy’s cat is not in the bedroom”), but he must also reconsider whether to trust his senses. Since perception is a fundamental source of information, an epistemically motivated agent can hardly postpone the decision whether to trust her senses for very long. The Umbrella case and the Cat case differ in some respects, but they seem to share two general characteristics. First, they begin with a similar problem: Anne and Andy must act in the context of uncertainty and pressure.40 Second, both cases can be reasonably resolved by following the pragmatic policy of avoiding greater harm: Anne can bring an umbrella to avoid (the greater harm of) getting soaked and becoming sick, and Andy can decide to trust his senses to avoid (the greater harm of) suspending judgment on (all) perception-based propositions. As a result, not only do the Umbrella case and the Cat case tackle a similar problem, but they might also include the same pragmatic policy to resolve it.41 In order to avoid greater harm, Anne and Andy are acting on presumptions. On a standard view, presumptions are propositions (or statuses of propositions) that agents proceed on in circumstances of uncertainty and pressure (Ullmann-Margalit 1983; Rescher 2006; Walton 2014; Godden 2017, 2019).42 In the Umbrella case, “It will rain” represents a practical presumption: although “It will rain” is uncertain, Anne acts on it by deciding to take an umbrella. In the Cat case, “Andy’s cat is in the tree” represents a cognitive presumption: although “Andy’s cat is in the tree” is uncertain from a sceptical viewpoint, Andy acts on it by relying on his senses and by using it as a tentative premise in reasoning. 40 

On the one hand, Anne must decide whether to take an umbrella although “It will rain” is an uncertain proposition.

On the other hand, Andy must decide whether to trust his perception and accept that “Andy’s cat is in the tree,” although neither perception nor the proposition is epistemically certain. 41 

One may object that the analogy does not hold entirely, since, ultimately, the Cat case is about believing rather than

deciding. Unlike Anne, Andy is an epistemic agent whose ultimate goal is to adopt justified beliefs, and belief-formation is hardly an instance of decision-making. Although I agree with this view, it must be stressed that, within the framework of epistemic pragmatism, the language of decision-making might still make sense in the context of epistemic inquiry. Of course, Andy is not supposed to decide whether he believes “Andy’s cat is in the tree,” but he might be able to decide whether it is reasonable to proceed on this belief and to derive tentative conclusions from it. Similarly, Andy is not supposed to decide whether he believes that his senses are reliable in general. Still, he might decide that it is epistemically reasonable to proceed on this belief, i.e., to adopt trusting perception as a default epistemic policy. So, in some sense, both the Umbrella case and the Cat case involve decision-making. I thank a reviewer of Argumentation for pressing this clarification. 42  On standard accounts, they are also closely related to the asymmetrical allocation of the burden of proof (see Whately 1963; Rescher 1977, 2006; Pinto 2001; Freeman 2005; Godden & Walton 2007; Walton 2014). However, the relationship between presumptions and burdens of proof is the topic of Chapters 3 and 4 (i.e., Bodlović 2020b; Bodlović, forthcoming).

52

On the Differences Between Practical and Cognitive Presumptions

For the most part, agents accept presumptions “to get on smoothly with business of all sorts, to cut through impasses, to facilitate and expedite action” (Ullmann-Margalit 1983, p. 147), but scholars agree that presumptions ultimately rest on normative foundations: [A]n important similarity of all of the accounts surveyed is that the foundations of presumptions are normative. Whether these foundations are explained in terms of institution-specific rules, general epistemic principles, the illocutionary consequences of making utterances of a certain kind, or the social obligations that envelop our day-to-day activities, presumptions require a grounding in norms. (Godden & Walton 2007, p. 337) The normative foundations seem compatible with pragmatic justifications of presumptions since, usually, pragmatic policies promote values. Anne, for instance, might decide to avoid greater harm, but what constitutes greater harm in her case depends on (her) preferences. She might choose to bring an umbrella because she values health more than convenience. There are also epistemic goals and preferences. For instance, after a sceptical remark, Andy might proceed as if his senses are reliable and as if “Andy’s cat is in the tree” is a justified proposition. This is because he would rather adopt (some) beliefs as true/justified at the expense of risking to accept some falsities, than avoid adopting (any) false/unjustified (perceptual) beliefs at the expense of failing to believe any truths. So, as far as kinds of goals, values, or preferences are concerned, there are practical and cognitive presumptions. If an agent proceeds on p to promote a non-epistemic goal and, typically, to avoid greater harm in circumstances of uncertainty and pressure, then p is a practical presumption. By contrast, if an agent proceeds on p to promote an epistemic goal and, typically, to avoid greater harm in circumstances of uncertainty and pressure, then p is a cognitive presumption. I adopt this terminology from Nicholas Rescher (2006, p. 27).

2

Practical and cognitive presumptions are different. Typically, they belong to different contexts (deliberation vs. inquiry) and have distinct ultimate contextual functions (arriving at decisions vs. acquiring true beliefs). Also, they promote different kinds of goals (non-epistemic vs. epistemic) and, consequently, have “foundations of qualitatively different types” (Godden & Walton 2007, p. 337). This is not a matter of dispute. But what about the similarities between the Umbrella case and the Cat case? Do practical and cognitive presumptions avoid greater harm in circumstances of uncertainty and pressure? Do they operate in sufficiently similar ways so that ‘uncertainty,’ ‘pressure,’ and ‘avoiding greater harm’ have comparable meanings? Are practical and cognitive presumptions the same tool used for similar pragmatic purposes in different contexts, or are they different tools, the similarities of which become less relevant once their particular applications are concerned? In short, once we set the contextual and qualitative considerations aside, do cognitive presumptions follow their own logic or are they, as Rescher suggests, “the epistemic analogue[s] of ‘innocent until proven guilty’” (2006, p. 23)?

53

CHAPTER 2

In this chapter, I will explore how practical and cognitive presumptions operate in dialogical contexts. I will argue that they are, in fact, different dialogical statuses for (at least) three reasons: typically, (1) they perform different dialogical functions, (2) they have structurally different pragmatic foundations, and (3) they cannot always be defeated by the same kinds of evidence. Thus, in the context of reasonable dialogue, practical and cognitive presumptions are differently analysed and evaluated, and thereby merit distinct treatment in argumentation theory. Although, at a high level of theoretical abstraction, they may share enough features to form the class of operators called ‘presumptions,’ the chapter focuses on dialogical differences rather than on conceptual similarities. Also, note that this article starts from standard understandings of the notions of practical and cognitive presumption (and, for the most part, does not seek to justify their adequacy) when comparing how these two kinds of presumptions fare within reasonable dialogues. The special attention is given to Rescher’s account and his view that practical and cognitive presumptions have very similar pragmatic foundations, i.e., that they both promote avoiding greater harm in circumstances of uncertainty. I first outline a standard dialogical approach to practical presumption (Section 2). After presenting its standard features, I focus on the notion of cognitive presumption (Section 3) and discuss how the two kinds differ (Section 3.2—3.4). In Section 4, I provide a summary of the most relevant conclusions, qualify these conclusions, and propose some guidelines for further research.

2 Practical Presumptions: The Standard View Philosophical scholarship offers many incompatible accounts of the nature, function, justification, and overall importance of presumptions. Nevertheless, the so-called practical characterization of presumptions, influenced mainly by legal scholarship, is well established within this fragmented picture.43 Let’s begin with two legal examples. The most famous example is the presumption of innocence. It is grounded in the rule in criminal law requiring that the accused should be considered innocent until or unless she is proven guilty. This presumption serves to resolve what Ullmann-Margalit (1983, p. 152) and Godden (2017, p. 505) call a “deliberation problem:” when it is (a) uncertain on grounds of (adequate) evidence whether the accused is innocent or guilty and (b) a legal decision needs to be made, we should “try to minimize the conviction of innocent persons, even at the cost of letting guilty persons go free [because] the former is judged the greater injustice” (Walton 1988, p. 244). As Dare & Kingsbury (2008, p. 507) put it, “better a hundred (in truth) guilty people go free than one (in truth) innocent person is jailed.” Another well-known example is the presumption of death, where the person who has been absent (without any explanation) for more than x years is presumed dead until proven otherwise. Although this presumption 43 

For an overview of various approaches to presumption in law, see Gama (2017). For a similar overview within

argumentation theory, see Godden & Walton (2007) and Lewiński (2017).

54

On the Differences Between Practical and Cognitive Presumptions

has some epistemic support (unlike the presumption of innocence), it is primarily a means to achieve a non-epistemic end: typically, it enables the distribution of the missing person’s estate when there is insufficient evidence indicating whether the person is dead or alive (Ullmann-Margalit 1983, p. 146; Rescher 2006, p. 27). Paradigmatic examples are useful for illustrating how practical presumptions typically operate. But what exactly are presumptions? What are their central features? The standard approaches define presumptions as appropriately qualified claims: proposition p counts as a presumption if and only if p is introduced (explicitly or implicitly) with the modal operator (status, qualifier) ‘presumably’ (see Ullmann-Margalit 1983; Hansen 2003; Rescher 2006; Godden & Walton 2007; Walton 2014; Godden 2017, 2019; cf. BermejoLuque 2016). Still, what does ‘presumably’ stand for?

2

The standard answer to the latter question places presumptions in a dialectical setting where parties exchange arguments in order to resolve a difference of opinion. Within this setting, the operator ‘presumably’ has unique deontic implications. On the one hand, the status of p as a presumption entitles the presumption’s User (proponent) to apply p in an argument without providing reasons for it: when p gets challenged, the User is not (immediately) obliged to argue in p’s favour. On the other hand, if the Challenger (opponent) is unwilling to accept p as a shared commitment, she is obliged to justify her rejection of p by providing reasons (see, e.g., Pinto 2001; Rescher 2006; Walton 2014; Godden 2017). Accordingly, the key feature of presumption is both dialectical and deontic: in a dialogue, ‘presumably’ indicates that some tentative commitment entails an asymmetrical allocation of the burden of proof. We may call this the deontic function of presumption. The practical presumption of innocence nicely illustrates this asymmetrical allocation: the defence is not obliged to prove the defendant’s innocence, whereas the prosecutor is ultimately obliged to prove the defendant’s guilt. This asymmetry is supposed to apply to cognitive presumptions, as well (Rescher 2006).

2.1 The Dialogical Function of the Practical Presumption With this characterization in place, what is the dialogical function of practical presumptions? By asymmetrically allocating the burden of proof, “[w]hat do presumptions do for the dialogue” (Bodlović 2017, p. 518)? How do they contribute to its constructive development and what dialogical goals are they supposed to achieve? What dialogical issues are they supposed to resolve? The paradigmatic dialogical function of practical presumptions is to enable dialogical progress in circumstances of pressure and uncertainty (see Ullmann-Margalit 1983; Walton 1988, 2008b, 2014; Godden 2017, 2019). Suppose that agents engage in an argumentative dialogue in order to resolve an urgent issue. Suppose that the deadline is approaching, that there is pressure to resolve the issue, and that the resolution depends on whether p is true.

55

CHAPTER 2

Suppose, however, that p is uncertain, i.e., that there is insufficient evidence to believe p.44 In these circumstances, we are facing the so-called deliberation problem and the obligation to provide sufficient reason for p will cause a dialogical deadlock to ensue. Since we cannot afford this due to urgency, we need an effective means to break the deadlock. The presumptive status of p is exactly that—it places the burden of proof on the opponent and allows us to proceed tentatively as if p were true. It enables the dialogue to progress, to continue on a provisional basis towards the resolution of an urgent issue.

2.2 Practical Presumptions and Presumptive Reasoning By asymmetrically allocating the burden of proof, practical presumptions enable the progress of deliberation. But what conditions warrant the acceptance of practical presumptions? Why exactly should we proceed as if p is true instead of presuming non-p and arriving at a different decision? What do the foundations of practical presumptions look like? In this section, I propose an answer by analysing the structure of presumptive practical reasoning. Inspired by legal scholarship, argumentation scholars typically reconstruct practical presumptions as conclusions of reasoning. This reconstruction reveals the logical function of presumptions: usually, practical presumptions (‘presumed facts’) are interpreted as conclusions of presumptive reasoning consisting of a ‘basic fact’ and a ‘presumptive rule’ (Ullmann-Margalit 1983; Hansen 2003; Rescher 2006; Godden & Walton 2007; Walton 2014; Godden 2017). The basic fact is an elementary or complex statement that gives rise to the presumption by instantiating the antecedent of the presumptive rule. The presumptive rule is a conditional that stands for a policy that prescribes the course of action (UllmannMargalit 1983; Rescher 2006). Defeasibility is a necessary feature of a presumptive rule.45 Although there are different ways to express defeasibility, I will simply use the ‘all else being equal’ phrase: a presumption follows from the basic fact, in accordance with the presumptive rule, if all else is equal, i.e., absent defeaters.46 44 

Legal dialogues provide good examples. After all, they cannot last forever: at some point, when all the available

(often inconclusive) evidence is taken into account, a decision must be made. 45 

Although some scholars, inspired by legal scholarship, acknowledge the existence of indefeasible, irrebuttable, or

conclusive presumptions (see Walton 2014, p. 107; Godden 2017, pp. 507-508), this chapter does not deal with such kinds. However, it is possible that “indefeasible ‘presumptions’ are presumptions in name only” (Rescher 2006, p. 5), since it is controversial whether they can place the burden of proof on the Challenger (Kauffeld 2003, p. 134). Also, it is important to keep in mind that defeasibility is not a sufficient condition to render some piece of reasoning presumptive. For instance, a statistical syllogism is surely a defeasible inductive argument, but it is (at least) debatable whether it represents a presumptive argument in the strict sense of the word. 46  Importantly, ‘all else being equal’ is not an element of a presumptive rule, but merely the indicator of its defeasibility. Some scholars express defeasibility in a different way, e.g., by adding the so-called ‘no-defeater clause’ as the second conjunct of a complex antecedent of some presumptive rule. What I call the ‘no-defeater clause’ has been labelled in various ways in the literature, such as, “rebuttal clause” (Ullmann-Margalit 1983, p. 149) and “default proviso” (Rescher 2006, p. 33). Although this indicates that one may make the presumption only in the absence of sufficient defeaters, treating

56

On the Differences Between Practical and Cognitive Presumptions

Although this explains why practical presumptions are not assumptions, presumptions still remain somewhat arbitrary. It is unclear why we should follow one presumptive rule rather than another. For instance, we can reconstruct the reasoning in the Umbrella case, as follows: 1. If rain is uncertain (basic fact), then, all else being equal, the deliberative agent should act on “It will rain.” [Presumptive rule] 2. The weather forecast website estimates that there is 50% chance of rain. [Basic fact] 3. Therefore, all else being equal, Anne should act on “It will rain” (i.e., “Presumably(p), it will rain’”) [Presumed fact/presumption]. But is there a reason why she should follow this presumptive rule? Why shouldn’t she start from the same basic fact and, due to insufficient evidence that it will rain, draw the contrary conclusion? After all, the basic fact “The weather forecast website estimates that there is 50% chance of rain,” together with the absence of defeaters, equally supports “It will rain” and “It will not rain.”

2

We can answer this question by expanding the core structure of presumptive reasoning. In the complete formulation, presumptive reasoning involves various considerations that (directly or indirectly) support the presumptive rule. Although basic facts (typically) and the absence of defeaters represent “the epistemic conditions under which [presumptive rules] come into effect” (Godden 2017, p. 506), presumptive rules are primarily supported by normative considerations. Thus, one can select the rule on grounds of safety, by appealing to the “principle of tutiorism” (Walton 1988, p. 247) or the “principle of precaution” (2014, p. 214). Of course, safety is one among many non-epistemic goals and values that can provide normative support for a presumptive rule. According to BermejoLuque (2016, p. 12), presumptive rules can also promote honesty and politeness, protect the value of human life, or increase the efficiency of some process/procedure. Crucially, when the ultimate normative goal is in place, one selects the presumptive rule in line with the policy of avoiding greater harm. This policy reveals the paradigmatic pragmatic (instrumental) function of presumptions: if there is an “expected utility imbalance with respect to p” (Aijaz et al. 2013, p. 270), i.e., if proceeding (erroneously) on p is less costly than proceeding (erroneously) on non-p, then one shall act upon p, or, in other words, presume that p. Accordingly, Anne should follow the presumptive rule that safeguards her actions in the special circumstances of risk and uncertainty. Let us spell out Anne’s reasoning in the Umbrella case more systematically. First, Anne realizes that two errors are possible: either she takes an umbrella and it does not rain, or she does not take an umbrella and it rains. Anne then estimates and compares potential the absence of defeaters as a premise in the antecedent, in fact, conceals the defeasible nature of presumptive reasoning. (It could not be the case that the premises are true, and that the presumption does not follow.) For this reason, I stick with the ‘all else being equal’ interpretation. I thank prof. James Freeman for explaining the advantage of avoiding reconstructions of presumptive reasoning in terms of the ‘no-defeater clause’ (see Freeman 2020).

57

CHAPTER 2

costs. The first error will cause only a slight inconvenience: Anne will have to carry around an umbrella without any need. The second error, however, might cause greater harm: Anne will probably get soaked and, in the worst-case scenario, she may become sick. Thus, Anne presumes “It will rain” and proceeds by taking an umbrella. Given that Jim shares Anne’s values and has a similar take on the risks involved, he should either adhere to Anne’s presumption or provide (additional) evidence that it will not rain. Let us now present the complete scheme of presumptive practical reasoning.47 Here, A stands for the deliberating agent (Anne); q stands for the basic fact (“The weather forecast website estimates that there is 50% chance of rain”); p (“It will rain”) and non-p (“It will not rain”) stand for propositions that can be acted upon; C1 (inconvenience of carrying an umbrella) and C2 (agent gets soaked and runs the risk of becoming sick) stand for the potential consequences of acting erroneously on either p or non-p; and G1 (health) and G2 (convenience) stand for basic goals (values) that underlie Anne’s deliberation. 1. Condition q obtains. (basic fact) 2. A is faced with a decision whether to act on p or non-p. (deliberation dilemma) 3. A cannot postpone the decision. (deliberation pressure) 4. A is not aware of sufficient evidence for either p or non-p. (genuine uncertainty)

evidential

5. If any deliberating agent is faced with a decision whether to act on some proposition, cannot postpone the decision, and is not aware of sufficient evidence for a proposition, then she is facing a deliberation problem. (deliberation problem) 6. If any deliberating agent is facing a deliberation problem, then she should seek to avoid a costlier error. (tutiorism, precaution) 7. If any deliberating agent acts on p and p turns out to be false, then C1 follows. (error-cost entailment 1) 8. If any deliberating agent acts on non-p and non-p turns out to be false, then C2 follows. (error-cost entailment 2) 9. If (a) G1 is more important than G2 to some deliberating agent; (b) C2 undermines G1, and (c) C1 undermines G2; then C2 is costlier than C1. (goal-cost entailment) 10. G1 is more important than G2 to A. (normative inequality) 11. C1 undermines G2. 12. C2 undermines G1. 13. C2 is costlier than C1. (cost comparison) 47 

[9, 10, 11, 12]

The following scheme is an expanded version of the “negative practical reasoning scheme,” proposed by Walton,

Reed and Macagno (2008, p. 100). Authors characterise the scheme as a “subspecies of the ad ignorantiam scheme” (p. 99).

58

On the Differences Between Practical and Cognitive Presumptions

14. A is facing a deliberation problem.



[2, 3, 4, 5]

15. A should seek to avoid a costlier error. (policy of avoiding costly errors) [6, 14] 16. All else being equal, following a presumptive rule that prescribes acting on p is a good means to avoid a costlier error. (expected utility imbalance) [7, 8, 13] 17. If some condition (such as q) obtains, then, all else being equal, the deliberating agent (such as A) should act on p. (presumptive rule) [15, 16] 18. Therefore, all else being equal, A should act on p (Presumably(p), p.) (practical presumption) [1, 17] Admittedly, this scheme renders justifications of practical presumptions more complicated than they intuitively seem. However, it is necessary to explain in some detail both the circumstances in which practical presumptions arise, as well as their pragmatic and normative foundations.

2

But is the concept of presumption necessary to explain or justify Anne’s decision to take the umbrella? It may seem that we can reconstruct her reasoning by taking into account only the probability of rain and the strength of her preferences to stay dry and travel light. If the expected utility of bringing an umbrella is greater than the utility of leaving it at home, then we can understand Anne’s decision without referring to the concept of presumption. Introducing presumptions into practical reasoning, then, might be an unnecessary complication that leads to undesirable consequences: for instance, Anne becomes (needlessly) forced to include an epistemically unwarranted proposition, namely the presumption “It will rain,” into her commitment set. Since this is an important objection, let me offer three clarifications.48 First, in my opinion, this objection might rest on the assumption that presumptions represent new propositions that we introduce next to other components of practical reasoning (such as preferences and policies). But this is hardly true. Even if we can explain and justify decisions without appealing to presumptions, this is not because we needlessly introduce presumptions in addition to other components of reasoning (such as preferences and policies), but because we introduce them by introducing other elements of reasoning. “Presumably(p), p” means that (1) proposition p (as well as non-p) is epistemically uncertain; that (2) proceeding on p avoids greater harm (3) given the agent’s goals; and that, for these reasons, (4) an agent takes on a tentative, practical commitment to proceed on p. So, technically, presumptions represent statuses of propositions, and these statuses seem inherent to practical reasoning. Even if presumptions are redundant, this is not because we introduce new propositions into reasoning and commitment sets out of nowhere, but because we can reduce presumptive status to more fundamental conditions of practical reasoning.49 48 

I thank a reviewer of Argumentation for raising this objection.

49 

For the argument that presumptions have an epiphenomenal character, i.e., that they are based on more fundamental

considerations, see Lewiński (2017).

59

CHAPTER 2

More importantly, even if presumptive status only designates a particular type of practical reasoning, mostly summarizes Anne’s reasoning or, perhaps, expresses Anne’s decision in different terms,50 it is still explanatorily relevant. Remember that the Umbrella case begins with Anne’s attempt to make an epistemically informed decision. Since the likelihood of rain motivates Anne’s reasoning in the first place, presumptive status is indeed relevant. That is, “Presumably(p), it will rain” explains the tentative and procedural nature of Anne’s commitment towards “It will rain.” Since other qualifiers (such as ‘plausibly,’ ‘probably,’ or ‘supposedly’) do not quite fit this kind of commitment, ‘presumably(p)’ is necessary to explain Anne’s relation to an important condition of her reasoning. Moreover, since we often analyse situations like the Umbrella case in terms of ‘presumptions,’ it is licit to make the role of presumptions visible in such reasoning (rather than covering it up). Finally, the procedural nature of ‘presumably(p)’ explains why introducing “It will rain” into Anne’s commitment set is not problematic. Anne’s acceptance of “Presumably(p), it will rain” is reasonable even though the proposition “It will rain” is epistemically unwarranted. So, the concern that adding “Presumably(p), it will rain” into a commitment set is problematic because “It will rain” is epistemically unwarranted does not seem plausible. In a practical case, ‘presumably’ is, by definition, a non-epistemic modality (Ullmann-Margalit 1983; Godden 2017, 2019).

2.3 What Can Defeat Practical Presumptions, and When? We have just seen how practical presumptions come to life. But how are they put to rest? Godden (2017) identifies four general defeating strategies: 1. The Challenger may criticize the tenability of any component that gives rise to presumptive status (such as the basic fact or the presumptive rule);51 2. The Challenger may undermine presumptive reasoning (by showing that the presumptive rule is not correctly applied on a given occasion or by introducing an undercutting defeater); 3. The Challenger may override presumptive reasoning (by questioning the User’s goal and, usually, proposing an alternative course of action based on different normative grounds); 4. Finally, the Challenger can rebut the conclusion “Presumably(p), p” by showing that p is (or could more likely be) false (see Godden 2017, pp. 506-507). 50  Somebody may claim that “Presumably(p), p” only means “We should act on p.” If this were true, then, technically, a decision would not be based on presumption, but it would be semantically equivalent to it. I disagree with this interpretation since, sometimes, we should act on p even when p is evidentially certain, i.e., when p is not a presumption. 51 

Here, Godden talks about the rebuttal of “the inferential conditions giving rise to the presumption” (2017, p. 506).

Although this is correct, I find it useful to make a terminological distinction between showing that the premise is false and showing that the ‘presumed fact’ is false. Hence, following van Laar & Krabbe (2013), I shall use ‘premise tenability criticism’ for the former and, following Pollock (1987), ‘rebuttal’ for the latter.

60

On the Differences Between Practical and Cognitive Presumptions

This list answers the question: What kinds of considerations, in principle, defeat practical presumptions? However, a related question has mostly escaped the attention of scholars, namely: When do these considerations exactly defeat practical presumptions? Once introduced into the dialogue, do they necessarily defeat presumptions (under the assumption that defeaters themselves remain undefeated)?52 Suppose, for instance, that the opponent challenges a presumption by introducing an undercutting defeater, and that the presumption’s User is committed to accept this defeater. Does the undercutting defeater necessarily cancel presumptive status? I will argue that, sometimes, given the particular background conditions, the undercutting defeaters are entirely useless for challenging practical presumptions, insofar as we assume the standard meaning of the notion of ‘undercutting defeater’ (see, e.g., Pollock 1987; Prakken & Sartor 2009; Krabbe & van Laar 2011). I will investigate this issue in some detail because I hypothesize that undercutting defeaters have different implications when they concern practical presumptions as opposed to cognitive presumptions.

2

According to Pollock’s (1987, p. 485) classic account, the undercutting defeater u is a piece of evidence that challenges the reliability of the connection between some premise q and some conclusion p. Although u renders the belief that p evidentially uncertain (unjustified), u is consistent with the premise q and the conclusion p. Suppose, following Pollock’s well-known example, that an object X appears red to you. Normally, the premise “X looks red to me” (q) epistemically justifies the conclusion “X is red” (p) but if, at some point, you learn that X is illuminated by red lights, then “X looks red to me” ceases to be a reliable indication that X is red. Although this new information is consistent with both “X looks red to me” and “X is red,”53 it forces an epistemic agent to suspend the belief (conclusion) that X is red. Hence, “X is illuminated by red lights” (u) is an undercutting defeater: by rendering the evidential connection between q and p unreliable, it makes the belief p epistemically unwarranted. To show that, sometimes, undercutting defeaters cannot defeat practical presumptions, let us go back to the Umbrella case. Although Anne presumes “It will rain” (p) due to “The weather forecast website estimates that there is 50% chance of rain” (q), we should not test the effectiveness of an undercutting defeater by applying it to the connection between q and p. Instead, we should focus on the connection between q and the belief that underlies the 52  Also, a third question arises: How do relevant considerations and strategies exactly defeat practical presumptions? Namely, presumptions can be defeated, or put to rest, in two distinct senses. First, the Challenger can cancel a presumption. This means that as soon as she puts forward a defeater, the proposition loses its presumptive status: what was a presumption prima facie, is not a presumption ultima facie. This sense of ‘defeated,’ I believe, nicely coheres with the premise tenability criticism, undercutting, and rebutting. Second, the Challenger can render a presumption inoperative, i.e., make it a pro tanto presumption. This means that the presumptive status might still have some weight, but it is set aside due to more important considerations. This sense of ‘defeated,’ I believe, nicely coheres with the overriding and, perhaps, undermining strategies. In this chapter, I focus on the cancellation of presumptions by means of an undercutting defeater. I thank a reviewer of Argumentation for pressing this clarification. 53 

It is possible that X is, in fact, red and also illuminated by red lights.

61

CHAPTER 2

presumption, i.e., Anne’s belief that there is 50% chance of rain.54 Hence, our question is the following: Given that the undercutting defeater u defeats the evidential connection between “The weather forecast website estimates that there is 50% chance of rain” (q) and the belief “There is 50% chance of rain” (r), does u always defeat “Presumably(p), it will rain” (p)? I believe that it does not. Suppose Jim discovers that, on April 1st, the weather forecast website is playing an April Fool’s joke: it estimates 50% chance of rain everywhere in the world. This information is an undercutting defeater: “The weather forecast website is playing an April Fool’s joke” (u) undercuts the reliability of the connection between q and r even though it is consistent with both q and r. When Jim puts forward an undercutter, Anne is no longer justified in believing that “There is 50% chance of rain” on the grounds that the weather forecast says so. But does Jim’s discovery defeat Anne’s practical presumption “It will rain,” and should Anne, given the joke, leave the umbrella at home? We can answer these questions only by taking into account particular background conditions related to the dialogue at hand, such as the prior likelihood attributed to “It will rain.” In the original version of the Umbrella case, Anne and Jim attribute equal likelihoods to “It will rain” and “It will not rain.”55 Under these circumstances, the information about an April Fool’s joke changes the nature of Anne’s epistemic position and, possibly, deepens her uncertainty: in some sense, Anne goes from being justified in believing that rain is uncertain to being less justified in believing so.56 But this change affects neither Anne’s decision to bring an umbrella nor the presumptive status of “It will rain.” Since the weather forecast is unreliable, and the prior likelihood of rain is 50%, nothing relevant has changed. Anne should still suspend judgment on “It will rain,” opt for a less costly mistake, presume that it will rain, and bring an umbrella. The undercutting defeater does what it is supposed to do, but the presumptive status remains in place.57 Under different circumstances, the undercutting defeaters can indeed cancel practical presumptions. Suppose that the prior likelihood of “It will rain” was sufficiently higher than the prior likelihood of “It will not rain.” When an April Fool’s joke shows the weather 54 

The reason lies in the tension between the procedural, non-doxastic nature of the operator ‘presumably(p)’ and the

natural target of an undercutting defeater. On the one hand, “Presumably(p), p” does not denote a belief that p, but a procedural attitude towards p (i.e., proceeding as if p is true). On the other hand, the target of an undercutting defeater is the evidential connection between some premise q and a doxastic attitude towards p, e.g., a belief that p. Therefore, testing the effectiveness of an undercutting defeater u by using it to bring into question the connection between a basic fact q and a practical presumption p probably leads to a categorical confusion. 55  That they attribute equal prior likelihoods to “It will rain” and “It will not rain” means that, in their opinion, prior to checking the weather forecast, the chance of rain in that geographical area is roughly 50% on a regular day of the season. For the sake of the example, we can assume that Anne’s and Jim’s estimations of likelihoods are correct. 56 

Typically, the belief “There is 50% chance of rain” is more epistemically justified when it is based on a reliable

weather forecast than when it is based only on prior likelihood of rain (even if the estimation of prior likelihood is correct). 57 

To be sure, u renders p evidentially uncertain, but we acknowledge the uncertainty of p by presuming p in the first

place. Practical presumptions are, by definition, tools for overcoming evidential uncertainty. Since we presume p due to its uncertainty, p can hardly lose its presumptive status due to its uncertainty.

62

On the Differences Between Practical and Cognitive Presumptions

forecast to be unreliable, the prior likelihood of rain becomes Anne’s only undefeated evidence. Thus, “It will rain” becomes Anne’s epistemic expectation and thereby loses its presumptive status, since practical presumptions are not epistemic expectations. In this case, the undercutting defeater u cancels p’s presumptive status by strengthening p’s epistemic status. Note, however, that Anne’s original decision should remain unchanged: since she expects rain, she should bring an umbrella. Finally, suppose that the prior likelihood of “It will rain” is sufficiently lower than the prior likelihood of “It will not rain.” When an April Fool’s joke shows the weather forecast to be unreliable, “It will not rain” becomes Anne’s epistemic expectation, and, based on this, she should decide not to bring an umbrella. Thus, “It will rain” loses its presumptive status. In this case, an undercutting defeater u eliminates p’s presumptive status by, ultimately, creating favourable conditions for an agent to rebut p, proceed on non-p, and make a different decision. The association between prior likelihoods and the effects of an undercutting defeater shows a different pattern than in the case of cognitive presumptions, as I will explain in Section 3.4.

2

In summary, Godden is right that “[practical] presumptions are defeasible in many of the usual ways” including “the discovery of undercutting defeaters” (2017, p. 506). But this does not mean that defeaters, introduced by challengers, necessarily defeat practical presumptions (assuming that the defeaters themselves remain undefeated). I argued that an undercutting defeater u cannot cancel a practical presumption p when the prior likelihood of p is the same as the prior likelihood of non-p (or when likelihoods are unknown). In this particular case, u creates conditions that, typically, bring practical presumptions to life rather than put them to rest.58 As we shall see, this is important for exploring the differences between practical and cognitive presumptions. Let us now examine the latter class of presumptions.

3 Cognitive Presumptions, and How They Differ from Practical Presumptions Although presumptions were originally analysed in the context of practical deliberation, we can also find them in epistemic contexts. Rescher and Freeman have been leading the way in the epistemic study of presumptions. Both scholars agree that typical cognitive presumptions arise from epistemic sources and establish tentative starting points in the dialogue. However, unlike Freeman, Rescher repeatedly stresses that presumptions are ultimately based on pragmatic policies of an epistemic nature (2006, p. xii; p. 38; p. 46; p. 48) that are, in turn, evaluated on ‘economic’ grounds—in terms of their epistemic costs and epistemic benefits (p. 54). This renders Rescher’s account of cognitive presumption much closer to the concept of practical presumption and, thereby, makes it a more suitable starting point for our inquiry in this section. 58  This is made transparent in our complete scheme of practical presumptive reasoning: evidential uncertainty (premise 4) is one of the key conditions of presumptive reasoning and represents a constitutive element of the deliberation problem (premise 5).

63

CHAPTER 2

3.1 Rescher’s Account of Cognitive Presumptions Rescher views cognitive and practical presumptions as sharing some essential features, even though their ultimate contextual functions are different. While practical presumptions guide “our decisions regarding actions,” cognitive presumptions are “made for the sake of answering our questions and filling gaps in our information” (p. 27). Thus, the two classes of presumptions belong to different contexts and serve different goals. Starting from this explanation, one may come to believe that cognitive and practical presumptions are materially rather than formally different. Contextual differences, by themselves, hardly indicate any difference in the formal conditions of presumptions, their justification or defeasibility, or the structure of their deontic properties. I will argue against this last proposal: there are significant structural differences between the two classes. For the most part, cognitive presumptions are tools that operate differently and come with a different instruction manual. What do cognitive presumptions look like? What cognitive policies do we have at our disposal? Here are two paradigmatic examples. First, we should trust our senses and memory. In the Cat case, Andy should proceed with his cognitive matters by taking “Andy’s cat is in the tree” as true until its presumptive status gets cancelled by a sufficiently strong defeater.59 Second, our prospects for acquiring information are better if we trust others. In the absence of definite proof, trusting people is simply a better cognitive policy than always doubting their competence, reliability, and honesty. Thus, if somebody asserts p, we should presume p and proceed with our cognitive matters unless we have good reasons to think that p is either false or unjustified. Trusting our senses and other people’s statements are, in the long run, economically rational policies: their cognitive benefits outweigh their cognitive costs (Rescher 2006, pp. 48-52). But what are cognitive presumptions? Rescher defines them as “truth-candidates, data that are no more certified truths than candidate-presidents are certified presidents” (2006, p. 37). However, the presumptions are not only truth-candidates but the most plausible truth-candidates: Presumption favors the most plausible of rival alternatives—when indeed there is one. This alternative will always stand until set aside (by the entry of another, yet more plausible, presumption). (Rescher 2006, p. 39) Two things are especially important here. First, according to Rescher, presumption is a “singulary” status (Freeman 2005, p. 26). This means that different cognitive rules may operate simultaneously and generate incompatible truth-candidates but, at each particular point, only the most plausible proposition becomes a presumption. Second, cognitive presumptions are defined in terms of ‘plausibility.’ This is a complex philosophical notion 59  By ‘proceeding with cognitive matters,’ I mean that Andy should feel free to derive (tentative epistemic) conclusions based on his presumption. For instance, he should freely derive “Andy’s cat is not in the bedroom at the moment.”

64

On the Differences Between Practical and Cognitive Presumptions

but, for our present purposes, it suffices to say that the degree of plausibility depends on the reliability of the source that vouches for a proposition (Rescher 1976, pp. 10-11). On the one hand, Rescher explains reliability in terms of “probative solidity,” “trustworthiness” and “authoritativeness” of the source (1976, pp. 6-7; 2006, p. 39). It is a broad construct that cannot be reduced to statistical considerations of a previous track record. On the other hand, he explains the concept of “source” by two different types of considerations: “evidentiation” and “principles” (2006, p. 40). To say that a proposition is evidentiated is to claim that it is prima facie supported by a standard epistemic source (in a narrow sense), such as sense-perception, memory, testimony, expert-testimony, or common knowledge. By contrast, principles render propositions plausible on grounds of simplicity, uniformity, or normality, and they come to the fore when presumptive status cannot be assigned on evidential grounds. Nevertheless, the usual and paradigmatic cognitive presumptions are based upon evidentiation. In the next sections, I will focus exclusively on the paradigmatic evidential presumptions and call them typical cognitive presumptions.

2

Before proceeding any further, we must note that cognitive presumptions are not the only means of promoting epistemic ends. Consider the case of two palaeontologists who arrive at different conclusions about the fate of Neanderthals (Elgin 2010, p. 54). Although the evidence is equivocal and the palaeontologists face a peer disagreement, suspending judgment would be very costly for scientific research. Thus, to facilitate inquiry about the fate of Neanderthals, both palaeontologists should proceed as if their conclusions were accurate, and further develop their positions (Elgin 2010, pp. 65-68). Although this case resembles cognitive presumptions in several ways,60 we should not portray p as a cognitive presumption for three different, but interrelated reasons. The first reason is conceptual: since p and non-p are equally plausible (evidentiated, simple, or normal), p cannot be the most plausible truth candidate. Second, since palaeontologists are encouraged to find and present reasons for their positions, the allocation of the burden of proof seems symmetrical. Finally, p and non-p are not dialectically privileged propositions: despite the fact that the evidence is equivocal (uncertain), the proponent of non-p (p) is not obliged to proceed as if p (non-p) is true. Thus, Elgin’s example reveals an important conceptual point. It shows that (1) proceeding on p (2) on defeasible grounds (3) to avoid greater harm (4) when p is uncertain (5) in the context of epistemic inquiry, is necessary, but not sufficient to render p a cognitive presumption (in Rescher’s sense of the term). In addition, p must be the most plausible truth candidate, and a dialectically privileged proposition that asymmetrically allocates the burden of proof.61

60  The palaeontologist will (1) proceed on p (2) in circumstances of uncertainty, and (3) will avoid suspending judgment on p in order to (4) avoid higher costs (5) in the context of epistemic inquiry. 61 

I thank a reviewer of Argumentation for pressing this clarification.

65

CHAPTER 2

3.2 What is The Dialogical Function of Cognitive Presumptions? How do cognitive presumptions distribute dialectical obligations? Do they asymmetrically allocate the burden of proof and, thus, have the same deontic function as practical presumptions? Rescher seems to think so. In his view, “burden of proof and presumption represent correlative conceptions inevitably coordinate with one another throughout the context of rational dialectic” (2006, p. 25). They are “opposite sides of the same coin” (p. 14). So why does Andy’s contention place the burden of proof on his Challenger? This is because we are naturally inclined to trust visual perception, as well as other people’s testimony (when circumstances are ordinary and the epistemic situation is simple). In everyday argumentation, we rarely doubt the reliability of standard epistemic sources. Thus, by refusing to concede “Andy’s cat is in the tree” on purely sceptical grounds, Jane is making an unusual move (both epistemically and dialogically): she is refusing to concede a highly plausible proposition for no case-specific reason. There is a strong intuition that, in ordinary circumstances, her move requires (additional) justification. By contrast, one may claim that Andy’s contention has presumptive status and that he should not be dialogically obliged to provide arguments (until Jane provides additional reasons for her case). In short, there is a presumption in favour of “what you may take for granted about a particular issue” (Cohen 1992, p. 13), or “how things ‘as a rule’ are taken as standing” (Rescher 1977, p. 30).62 But how does putting dialogical pressure on the Challenger contribute to the constructive development of the dialogue? What dialogical goals are cognitive presumptions supposed to achieve and what problems are they supposed to resolve? The dialogical function of cognitive presumptions is closely linked to the place they occupy in the structure of the reasonable dialogue. That is, scholars usually interpret cognitive presumptions as dialogical starting points: a set of shared premises tentatively accepted by (reasonable) interlocutors. This interpretation is proposed by Rescher (1977, 2006), pragmadialecticians (van Eemeren & Houtlosser 2002), Freeman (2005), van Laar & Krabbe (2013), and, occasionally, Walton (2014). All these scholars believe that presumptions are available to interlocutors from the very start of the argumentative exchange. 62  There are some objections to this view. Suppose that, indeed, cognitive presumptions represent a conservative force and serve to ‘dialectically protect’ the status quo. Next, suppose that the presumptive rule “We should trust experts” is in line with the status quo because, from the God’s eye view, people usually trust experts. Even in these circumstances, it is controversial whether presumptions are applicable in particular dialogical situations. First, “in actual practice people often tend to disagree as to what exactly is to be considered as the ‘status quo’” (van Eemeren & Houtlosser 2002, p. 19). The Challenger might not accept “We should trust experts” as an adequate presumptive rule and might insist that the burden of proof rests with the User of a presumption. Second, parties may agree about the status quo but disagree about its normative significance. For instance, the Challenger may believe that we should not derive normative conclusions from the fact that people usually trust experts (“is” does not imply “ought”). Also, an opponent may ascribe some normative significance to the status quo, but maintain that more significant considerations place the burden of proof on the User. Suppose that the Challenger believes that trusting experts is not as important as being epistemically independent and autonomous. Admittedly, these are all open questions, and they are motivated by serious and pressing issues. However, the complex analysis of the applicability of presumptions (and the various kinds of meta-discussions concerning the allocation of the burden of proof) falls outside of the present chapter’s scope.

66

On the Differences Between Practical and Cognitive Presumptions

Practical presumptions are different in these respects. In Godden’s view, they are not “the inferential resources already at hand” but rather “additional inferential capital” or the “new intellectual resources” used in order to “proceed with our undertakings” (2017, p. 487). This picture is fully compatible with the view that (many) practical presumptions are not dialogical starting points and, thereby, should not be part of the opening stage of deliberation. They may (also) come in handy during the later stages of dialogue (perhaps at the argumentation stage) when the evidential resources cease to provide guidance for reasonable decision-making. Since the two classes of presumption can belong to different dialogical stages, we can expect that they can also serve different dialogical functions. Rescher occasionally claims that cognitive presumptions enable dialogical progress; one may come to believe, then, that the two classes of presumptions have a comparable dialogical function. To see why this is not likely the case, let us examine the following quote:

2

There must clearly be some class of claims that are allowed to be at least provisionally accepted within the framework of argumentation, because if everything were contested, the process of inquiry could not progress at all. (Rescher 2006, p. 24) What Rescher means by ‘enabling dialogical progress’ is preventing the well-known problem of dialectical regress. Dialectical regress, which has intrigued philosophers ever since Pyrrhonian scepticism, results from the proponent’s inability to defend her standpoint against the ‘persistent interlocutor.’63 The persistent interlocutor is the opponent who automatically challenges every reason offered by the proponent (without offering anything in return). Suppose that the dialectical rule concerning the burden of proof is universal, namely, that every proposition introduced can be challenged and, if challenged, needs to be defended by the proponent. This allows the persistent interlocutor to sabotage the proponent’s aim of rationally persuading her by challenging the proponent’s claims ad infinitum. In principle, this situation can happen with both parties playing by the dialectical rules. An apparent solution is to change the rules. In well-known fashion of dialogical foundationalism, Rescher contends that “[w]e cannot argue everything discursively ‘all the way down’” (2006, p. 50). The burden of proof rule makes sense only if it allows for exceptions, i.e., if some propositions do not require defence once they are contradicted (see Rescher 1977, p. 33; 2006, p. 30). Since “[t]hey furnish a starting point” (2006, p. 50), this is where cognitive presumptions can help. By placing the burden of proof on the Challenger, they make the User immune to the Challenger’s unusual challenge and prevent the dialogue from collapsing into an endless chain of reasons and challenges. This contribution to dialogue is in many respects different from that of practical presumptions. Cognitive presumptions are often portrayed as dialogical tools for fighting scepticism (see Freeman 2005; Rescher 2006; Rescorla 2009b) and they seek to resolve a problem that, in a literal reading, does not arise in practice. Namely, as finite beings, we will hardly ever meet an interlocutor that challenges our reasons ad infinitum. A more 63  The term originally belongs to Adam Leite. I borrow it from Rescorla (2009b, p. 47).

67

CHAPTER 2

charitable interpretation, however, sees dialectical regress as related to a type of dialogue where the interlocutor automatically requests justification for every new reason put forward by the proponent, without offering anything in return. Thus, although the interlocutor does not challenge the proponent’s claims ad infinitum, she is still too persistent for the circumstances at hand. Cognitive presumptions can prevent persistent interlocutors from winning the argument by using this annoying strategy. That is, once the User introduces a cognitive presumption, the Challenger cannot request an argument without offering reasons in return. Thus, cognitive presumptions boost our immunity against a persistent interlocutor. In addition to being an ‘active dialogical cure,’ I believe that cognitive presumptions are also a normative ‘means of prevention:’ by limiting the potential of the persistent interlocutor’s strategy, cognitive presumptions may discourage many interlocutors from becoming persistent in the first place, thereby preventing the dialectical regress from ever occurring. By contrast, practical presumptions resolve problems only when they, in fact, arise. They do not spare us the trouble of making decisions under uncertainty but rather provide default solutions when we find ourselves in this kind of trouble. To stick with the metaphor, practical presumptions are a cure for typical deliberation problems.64 Of course, cognitive presumptions have a more mundane dialogical function. As (supposedly) mutually acceptable premises, they surely enable dialogical progress. But note that their role must still be different from that of practical presumptions. Practical presumptions move dialogues forward precisely when available (evidential) premises, including typical cognitive presumptions, cannot move them forward: they enable dialogical progress precisely when cognitive presumptions do not (see Bodlović 2017, p. 522). Accordingly, the two types of presumptions still face different dialogical issues, enable different kinds of progress, and have distinct dialogical potentials.65 In summary, although both classes of presumptions enable constructive development of the dialogue, they are standardly concerned with different dialogical problems (progress vs. regress) and, thus, usually do not have the same dialogical function. Also, they seem to represent different kinds of dialectical remedy: while practical presumptions are an active dialectical cure, cognitive presumptions are also a means of dialectical prevention. As far as the structure of dialogue is concerned, the two types of presumptions seem to be looking in different directions: typical cognitive presumptions provide an initial common 64  In many cases, cognitive presumptions stop the regress when the challenge is already under way, and the Challenger is already too persistent. Therefore, they can function as a cure, as well as a means of prevention. By contrast, paradigmatic practical presumptions seem to function only as a dialogical cure. 65 

Of course, this conclusion applies only to paradigmatic practical presumptions and typical cognitive presumptions.

Cognitive presumptions based on principles (such as simplicity, uniformity, etc.) can move dialogues forward when progress cannot be achieved on evidential grounds: for instance, we can decide to proceed on a simpler or more elegant proposition. Practical presumptions and cognitive presumptions based on principles are similar in this respect. This chapter, however, primarily focuses on typical cognitive presumptions. I thank a reviewer of Argumentation for pressing this clarification.

68

On the Differences Between Practical and Cognitive Presumptions

ground for the dialogue (and provide materials for it to commence and develop) whereas practical presumptions allow the dialogue to proceed towards its conclusion. Additionally, these differences might indicate that, typically, practical and cognitive presumptions do not belong to the same stages of critical discussion.

3.3 Cognitive Presumption and Presumptive Reasoning As far as logical function is concerned, cognitive presumptions (just like practical ones) are tentative conclusions derived from basic facts and defeasible presumptive rules (Rescher 2006, p. 33). Rescher emphasizes that the presumptive rule is precisely that—a rule, an imperative, an instruction on how to proceed with our cognitive matters. Once we establish a basic fact, the presumptive rule prescribes a particular action (acting, in the cognitive domain, as if a particular claim is true) based on “an epistemic policy [that] is closely analogous to the prudential principle of action—that of opting for the available alternative from which the least possible harm can result” (Rescher 2006, p. 39). Since, according to Rescher, cognitive and practical presumptive rules share a very similar pragmatic rationale (avoiding serious harm), we might expect that the two types of presumption share a closely similar pragmatic function.

2

Rescher’s pragmatic interpretation is theoretically promising,66 but to see how the pragmatic rationale works in cognitive contexts, we should test it in a concrete case. Does the Cat case confirm that the presumptive rule (of trusting one’s senses) is pragmatically justified, and is this justification analogous to the policy of ‘avoiding costly errors’ in the practical Umbrella case? In the remainder of this section, I will treat the complete scheme of presumptive practical reasoning as a template for reconstructing the complete scheme of cognitive reasoning. Despite efforts to save the analogy by preserving as many elements of the practical scheme as possible, the presumptive cognitive scheme appears different in several respects. So, what does pragmatic reasoning in the Cat case look like? First, Andy takes Jane’s sceptical remark seriously and deliberates on whether to accept “Andy’s cat is in the tree.” He identifies two potential errors: either he will falsely accept “Andy’s cat is in the tree” or he will adopt a sceptical stance towards a true (justified) proposition. Andy realizes that the second error will lead to more serious consequences. Adopting a sceptical stance towards this proposition for no case-related reason would mean that he no longer trusts his senses and, consequently, must suspend judgment on other empirical propositions. For a responsible epistemic agent, this is way too costly: suspending judgment on all empirical propositions, many of which are true (justified), is surely costlier than adopting some false (unjustified) beliefs. Thus, Andy decides to err on the side of a lesser evil, presumes that his cat is (in fact) in the tree and, as a matter of cognitive policy, continues to trust his senses. 66 

For instance, if both types of presumptions share a closely similar pragmatic rationale, then the notion of

‘presumption’ preserves conceptual cohesion. Also, the pragmatic justification of cognitive presumptions might provide a distinct methodological solution of the infinite dialectical regress (see Rescher 2006, pp. 57-59).

69

CHAPTER 2

Let us now construct the complete scheme of presumptive cognitive reasoning. In the scheme below, A stands for the cognitive agent (Andy); p (“Andy’s cat is in the tree”) and Sp (suspending judgment on “Andy’s cat is in the tree”) stand for attitudes that can be acted upon; q stands for the basic fact (“Andy sees that a cat, which looks exactly like his cat, is in the tree”); C1 (adopting some false or unjustified propositions) and C2 (suspending judgment on all true or justified empirical propositions) stand for potential consequences of acting erroneously on either p or Sp; and, finally, G1 (acquiring true or justified beliefs) and G2 (avoiding false or unjustified beliefs) stand for basic epistemic goals that underlie Andy’s deliberation. 1. Condition q obtains. (basic fact) 2. A is faced with a decision whether to act on p or Sp. (cognitive dilemma) 3. A cannot postpone the decision. (cognitive pressure) 4. A is not aware of sufficient (conclusive) evidence for either p or non-p. (academic uncertainty) 5. If any cognitive agent is faced with a decision whether to act on some proposition, cannot postpone the decision, and is not aware of sufficient evidence for a proposition, then she is facing a deliberation problem. (deliberation problem) 6. If any cognitive agent is facing a deliberation problem, then she should seek to avoid a costlier error. (tutiorism, precaution) 7. If any cognitive agent acts on p and p turns out to be false, then C1 follows. (errorcost entailment

1)

8. If any cognitive agent acts on Sp (either adequately or erroneously), then C2 follows. (error-cost entailment 2) 9. If (a) G1 and G2 are equally important to some cognitive agent, (b) C2 completely undermines G1, and (c) C1 undermines G2 to a limited extent; then C2 is costlier than C1. (goal-cost entailment) 10. G1 and G2 are equally important to A. (normative equality) 11. C1 undermines G2 to a limited extent. 12. C2 completely undermines G1. 13. C2 is costlier than C1. (cost comparison)

[9, 10, 11, 12]

14. A is facing a deliberation problem.

[2, 3, 4, 5]

15. A should seek to avoid a costlier error. (policy of avoiding costly errors) [6, 14] 16. All else being equal, following a presumptive rule that prescribes acting on p is a good means to avoid a costlier error. (expected utility imbalance) [7, 8, 13] 17. If some condition (such as q) obtains, then, all else being equal, the cognitive agent (such as A) should act on p. (presumptive rule) [15, 16] 18. Therefore, all else being equal, A should act on p (Presumably(c), p.) (cognitive presumption) [1, 17] 70

On the Differences Between Practical and Cognitive Presumptions

The scheme reveals that presumptive cognitive reasoning remains different from presumptive practical reasoning even when we reconstruct it with the error-management policy in view. In the following sections, I identify three (potential) differences: differences concerning the nature of the deliberation problem (Section 3.3.1), the difference concerning error-cost dependence (Section 3.3.2), and the difference concerning goal-cost entailment (Section 3.3.3). Although these differences still allow for some similarity between the foundations of practical and cognitive presumptions, they pose questions about strength and limits of similarity, and explain why Rescher’s position should be qualified.

3.3.1 Is There a (Comparable) Deliberation Problem?

2

The deliberation problem is the driving force behind practical presumptions, and it consists of three components: an agent faces (1) a deliberation dilemma, (2) (deliberation) pressure, and (3) (genuine) evidential uncertainty. But are practical and cognitive agents faced with closely similar deliberation problems? Does presumptive cognitive reasoning operate in an identical set of conditions as presumptive practical reasoning? Let us begin by discussing two minor differences. First, the foundations of practical and cognitive presumptions can include different deliberation dilemmas. In the practical case, Anne must decide whether to act on p (“It will rain”) or its negation non-p, and since she cannot act upon suspended judgment, the dilemma is perfectly reasonable.67 In the cognitive case, however, Andy can proceed on suspended judgment (despite the costs) and this sceptical response, rather than proceeding on non-p, constitutes the second horn of his dilemma. Thus, the two types of presumptive reasoning may involve a distinct set of assumptions and, thus, can rest on different deliberation problems. Second, cognitive and practical presumptive reasoning rest on distinct types of ‘pressure.’ Deliberation pressure is generated by the approaching deadline, while the urgency of making a timely decision is neither necessary nor typical for cognitive presumptions.68 Cognitive pressure, at least in the Cat case, is generated by the fact that the perception is a fundamental, unavoidable, and presumably irreplaceable source of empirical knowledge. So, cognitive pressure is created by our internal cognitive constraints, rather than external time constraints. Also, deliberation and cognitive pressures, for the most part, push agents towards different types of decisions. What is essentially at stake in a practical case is whether to proceed on a particular proposition at a particular time: whether Anne should proceed on “It will rain” when she is about to leave the apartment. What is essentially at stake in a cognitive case is whether to proceed on the source of the proposition, as a matter of default policy: whether Andy should always proceed by trusting 67 

What exactly would it mean to act on suspended judgment in the Umbrella case? That Anne takes only one half of

an umbrella? 68  Sometimes, the cognitive agent might find herself in similar circumstances. If the neighbour knocks on the door (just as Andy is contemplating about Jane’s sceptical remark) and asks about the cat’s whereabouts, this will surely put some external pressure on Andy to make a quick decision on how to proceed. This situation, however, is neither necessary nor typical.

71

CHAPTER 2

his perception in the absence of defeaters. These differences require further examination, but they already cast doubt on whether the two types of presumptive reasoning include analogous deliberation problems. The following, third difference is especially important in view of the nature of deliberation problem: usually, cognitive presumptions are not triggered by genuine evidential uncertainty. Although Rescher (2006) suggests that they are “low-grade data” (p. xii) that arise in “situations of incomplete information” (p. 37) and operate “in the region of uncertainty” (p. 166), he only means that cognitive presumptions lack conclusive support.69 By presupposing this conclusive standard of evaluation, Rescher renders cognitive presumptions uncertain, but this academic uncertainty is quite different from genuine evidential uncertainty (often associated with practical presumptions). This calls for the first, weaker conclusion: practical and cognitive presumptions are different because their foundations (often) include different kinds of evidential uncertainty and, thus, different kinds of the deliberation problem.70 However, we may also propose a stronger conclusion. Recall that Rescher defines cognitive presumptions as the most plausible truth candidates. Since this renders cognitive presumptions inconclusive by definition, suggesting that they are, in some sense, epistemically deficient for not meeting the conclusive (academic) standard of certainty comes close to committing a categorical mistake. Cognitive presumptions should be evaluated by standards of plausibility, and once this is done, they often cease to be evidentially uncertain. In fact, typical cognitive presumptions, as the most evidentiated truth candidates (see Rescher, 2006, p. 37) might call for the opposite epistemic evaluation: given that they belong to the nondeductive region of cognition, “the factual area where defeasibility is virtually inescapable” (p. 64), typical cognitive presumptions are often as certain as they can possibly be.71 The stronger conclusion, therefore, is the following: practical and cognitive presumptions are different because, typically, cognitive presumptions are not evidentially uncertain to begin with. This implies that typical cognitive reasoning, unlike the paradigmatic practical one, does not involve a deliberation problem. 69  They are not “outright” (Rescher 2006, p. 32) or “certified truths” (p. 28) that come with “categorical guarantees” (p. 31) or hold “with categorical assurance” (p. xi). They are not “absolutely certain or totally self-evidencing theses” (p. 20), and do not belong to Cartesian-style epistemology of certain knowledge (p. xiii). 70  Another way of showing that practical and cognitive presumptions are not uncertain in the same sense is to examine their basic facts closely. Basic facts in the cognitive scheme usually provide plausible epistemic grounds: it is epistemically plausible to presume “Andy’s cat is in the tree” on the grounds that “I see that a cat, which looks like Andy’s cat, is in the tree.” In the practical scheme, by contrast, basic facts do not typically provide reliable epistemic support: the practical presumption “It will rain” (or “John is innocent”) hardly follows from the basic fact “The weather forecast estimates that there is 50% chance of rain” (or “John is accused”). We may express this difference in structural terms by saying that, usually in practical presumptive reasoning, basic facts are covert ignorance premises (e.g., “It is not proven that it will not rain,” “It is not proven that John is guilty”). Practical presumptions, unlike typical cognitive ones, are typically based on arguments from ignorance. For a closer examination of the relationship between presumptions and arguments from ignorance, see Bodlović (2019b). 71  This interpretation, I believe, explains the fact that scholars describe cognitive presumptions as dialogical (epistemic) starting points more successfully than the alternative, ‘academic’ interpretation.

72

On the Differences Between Practical and Cognitive Presumptions

The stronger conclusion raises pertinent questions. Why should we even consider the policy of avoiding costly errors? Without evidential uncertainty and a deliberation problem, the motivation for accepting this policy is unclear, and we should try justifying its acceptance on alternative grounds. But is it reasonable to search for alternative grounds? If typical cognitive presumptions usually have solid evidential support, and this support renders errors unlikely, it seems unreasonable to reconstruct their foundations in terms of errors.72 Moreover, perhaps appealing to any policy is needless to begin with. Why shouldn’t accepting the presumptive rule “If p is the best evidentiated belief, then take p as true” rest on the simple fact that the best evidentiated belief is probably true or that “justified belief has an intimate link to true belief” (Goldman 2003, p. 62)? What does the reconstruction in terms of policy add to the latter interpretation, and what stops this reconstruction from ultimately becoming “just a piece of practical advice in accordance with some canons of inductive reasoning” (Ullmann-Margalit 1983, p. 158), which, then, “requires [no] special theory of presumptive inference” (Godden 2017, p. 498)? As we can see, the stronger conclusion has far-reaching consequences for the proposed account of presumptive cognitive reasoning.

2

In sum, the two schemes of presumptive reasoning differ from one another either because they involve different versions of the deliberation problem (due to different kinds of dilemmas, pressures and uncertainties) (weaker conclusion), or because the cognitive scheme does not involve any deliberation problem to begin with (stronger conclusion). The stronger conclusion could reveal further differences: the policy of avoiding costly errors is either inapplicable to cognitive presumptive reasoning, or its justification (and motivation) must differ from the paradigmatic practical case.

72  Admittedly, Rescher can insist that the justification of presumptive rules is not about track records and probabilities and that, accordingly, the low probability of error is irrelevant when discussing the foundations of cognitive presumptions: The justification of these presumptions is not the factual one of the substantive generalization. “In proceeding in this way, you will come at correct information and will not fall into error.” Rather, it is methodological justification. In proceeding in this way, you will efficiently foster the interests of the cognitive enterprise; the gains and benefits will, on the whole, outweigh the risks, losses, and costs. (Rescher 2006, pp. 49-50) But this only raises further questions. First, this kind of response would seem to depend on a sharp distinction between factual and methodological justification, which is rather controversial. Namely, how can following a presumptive rule “efficiently foster the interests of the cognitive enterprise” (pp. 49-50) if this rule is not effective for finding out truths and avoiding errors? In other words, how can cognitive benefits “on the whole, outweigh the risks, losses, and costs” (p. 50) if a cognitive rule makes errors probable and truths improbable? It seems that Rescher’s methodological justification might ultimately depend on factual (probabilistic) justification, and he seems to admit this. In the so-called cyclic revalidation of presumptions, “the appropriateness of implementing the principle in this or that case will depend critically on the overall track record of our experience of its use in comparable cases” (2006, pp. 58-59). So, our initial question remains reasonable: if the probability of error is low, why should the concept of error play a central role in the reconstruction of cognitive reasoning?

73

CHAPTER 2

3.3.2 Does the Cost of Action Depend on Proceeding on an Error? There might be another reason to doubt that the policy of avoiding costly errors is applicable to presumptive cognitive reasoning. Suppose that, in some cognitive cases, acting on a particular attitude, when this attitude is wrong, entails the same cost as acting on it, when it is right. If such cases exist, then, sometimes, cognitive costs do not genuinely depend on errors and, once again, we face the question: What justifies accepting the policy of avoiding costly errors in presumptive cognitive reasoning? Let me briefly explain what is meant by this error-cost independence. Consider two variations of the Cat case. In both variations, Andy sees that a cat is in the tree, Jane remarks that perception is generally fallible, and Andy suspends judgment based on her remark. Unfortunately, in the first scenario, it is both true and justified to believe that Andy’s cat is in the tree. Thus, Andy erroneously proceeds on suspended judgment and faces an enormous cost: due to the consistency requirement, he becomes obliged to suspend judgment on all perception-based propositions. In the second scenario, however, Andy is in luck. Since it is now both false and unjustified to believe that his cat is in the tree, suspension of judgment is (mostly) an adequate response. Crucially, despite his epistemic luck, the final cost remains the same: due to the consistency requirement Andy again becomes obliged to suspend judgment on all perception-based propositions. The cost does not seem to depend on making an error. Naturally, one might object that we are focusing on the wrong error. Suspending judgment solely on the grounds of perception’s fallibility is what makes Andy’s response problematic, and he pays the same price simply because he makes the same error in both scenarios: his mistake does not concern ‘one-time choice’ of a particular proposition but his general choice of cognitive policy. Although this is true, the previous point still contributes to exploring the differences between cognitive and practical presumptive reasoning. Namely, when suspending judgment on “Andy’s cat is in the tree” is adequate and Andy suspends judgment, he still bears the cost of not acquiring new perceptionbased propositions.73 By contrast, when “It will not rain” is true and Anne accepts this proposition, she does not bear the cost of getting soaked and running the risk of becoming sick. As a result, in some cases of presumptive cognitive reasoning, contrary to cases of practical reasoning, the cost of action does not depend on whether the agent’s action is erroneous or not.

73 

We might think of various reasons why Andy mistakenly sees his cat, from taking hallucinogenic drugs and

showing signs of mental illness, to confusing his cat with the neighbour’s cat. The crucial point, however, is that Andy does not suspend his judgment due to any case-specific reason, but due to Jane’s sceptical remark about the general fallibility of perception. Thus, by suspending judgment on “Andy’s cat is in the tree” due to a sceptical remark, Andy incurs a commitment to treat the fallibility of perception as a sufficient reason to suspend judgment on perception-based propositions. Due to the consistency requirement, Andy is now obliged to suspend judgment on every perception-based proposition.

74

On the Differences Between Practical and Cognitive Presumptions

That cognitive and practical reasoning are different in this respect is made transparent in reasoning schemes. In a practical scheme, C2 follows if the agent erroneously acts upon non-p (‘error-cost entailment 2’). By contrast, in a cognitive scheme, C2 sometimes follows regardless of whether the agent’s action on Sp is erroneous or not (‘error-cost entailment 2’). As a result, the two schemes of presumptive reasoning involve distinct assumptions. In the following subsection, I shortly discuss another potential difference between practical and cognitive schemes: the estimation and comparison of costs might depend on different normative conditions.

3.3.3 Does the Cost Depend on the Goals of Distinct Normative Importance?

2

What renders the consequence of one action less costly than the consequence of another action? On what grounds is C1 judged less costly than C2 in practical and cognitive presumptive reasoning? These questions surface yet another potential difference: practical and cognitive reasoning may involve different (normative) considerations when it comes to estimating, as well as comparing the costs of possible actions. Let me illustrate this in examples. In the Umbrella case, Anne estimates and compares the costs of two potential consequences—carrying an umbrella (when it does not rain) and getting soaked/ becoming sick—while fostering two practical goals of unequal importance (health trumps convenience). Since carrying an umbrella (when it does not rain) undermines a less important goal (convenience) but not a more important goal (health), Anne concludes that it is the less costly consequence of the two. Therefore, in paradigmatic practical cases, costs seem to depend on the normative inequality of underlying goals and values. In the Cat case, Andy estimates and compares the cost of two potential consequences and, similarly to Anne, fosters two underlying goals: acquiring true/justified beliefs and avoiding false/unjustified beliefs. The potential difference, however, is that these goals might be of equal epistemic importance (normative equality). Unlike health and convenience, acquiring truths and avoiding falsities might be two sides of the same coin: since we cannot promote knowledge without acquiring truths or by adopting falsities, acquiring truths and avoiding falsities are equally important epistemic goals.74 However, without the normative asymmetry between goals, undermining the first goal will entail, all else being equal, the same cost as undermining the second: accepting the false negative (rejecting p when p is true/justified) will have the same epistemic cost as accepting the false positive (accepting p when p is false/unjustified) (see, e.g., Pigliucci & Boudry 2014). Accordingly, the estimation of cognitive costs is likely to be different from the estimation of practical costs. Presumptive practical reasoning operates in the context of goals of unequal normative importance, where higher cost is derived from promoting a 74 

It is even possible to combine them into one unique goal which generates epistemic norms. For instance, Goldman

claims that “rules of good argumentation have the primary goal of promoting true belief and error avoidance” (1994, p. 29, emphasis added).

75

CHAPTER 2

less important goal. Presumptive cognitive reasoning operates in the context of equally (or at least comparably) important goals, where higher cost is derived from giving unfair treatment to equally important goals. A failure to treat epistemic goals fairly is precisely what renders the policy of suspending judgment epistemically inadequate. This policy avoids the risk of acquiring false/unjustified beliefs but, insofar as acquiring true/justified beliefs is concerned, it has catastrophic consequences. Of course, Rescher recognizes this: [I]t is always in our interest to proceed in ways that are efficient and effective in meeting our informational requirements. If playing safe were all that mattered, we would, of course, suspend judgment indefinitely. But it is simply not in our interest to do so, since safety is not all. A policy that would deprive us of any and all communicative benefit is inherently counterproductive. In rational inquiry we seek not only to avert error but also to achieve knowledge. (Rescher 2006, p. 122) Trusting perception represents normatively balanced policy. Although it promotes the goal of acquiring true/justified beliefs, it also instructs us to accept perceptual beliefs prima facie, i.e., only in the absence of sufficient defeaters. By recognizing the role of epistemic defeaters, the policy of trusting perception becomes consistent with the goal of avoiding false/unjustified beliefs. With this in mind, the costs of suspending judgment and trusting perception cannot be even: while the first policy leads to complete distrust, the second policy does not lead to gullibility. The policy of trusting perception treats goals of comparable epistemic importance more suitably and, thus, represents a less costly epistemic policy.75 Hence, while we do not wish to assign equal relevance to goals of unequal importance when estimating practical cost, we also do not wish to assign unequal relevance to goals of equal importance when estimating cognitive cost.76 The schemes of practical and cognitive presumptive reasoning make this transparent because they include different normative assumptions (normative inequality vs. normative equality) and different ‘goal-cost entailment’ premises. 75 

An alternative explanation of the difference might be that epistemic goals, unlike practical ones, are mutually

dependent. While health is at least not necessarily linked to convenience, acquiring truths is both conceptually and epistemically linked to avoiding falsities. Conceptually, the notion/value of truth seems to underlie both epistemic values and, epistemically, undermining the avoidance of falsities will systematically undermine the acquiring of truths (e.g., false beliefs will act as misleading defeaters for otherwise plausible inferences from true premises to true conclusions). This interpretation retains the possibility that the goals of acquiring truths and avoiding falsities, in some situations, have different weights. For instance, acquiring truths might be more important than avoiding falsities due to some nonepistemic, practical reasons: if you passionately care about whether p is true, and this knowledge comes at the cost of acquiring many false beliefs (for example, because you must be exposed to a source that is very unreliable for acquiring other beliefs), you may find it reasonable to meet this cost. But I am not sure if this can be reasonable from a purely epistemic perspective. Also, it is crucial to note that, although it rests on different premises, this alternative explanation still supports the conclusion that the cognitive policy should properly balance two epistemic goals. 76  For further and more subtle discussion of the distinct features of goals and values in practical and epistemic reasoning, see, e.g., Raz (2009).

76

On the Differences Between Practical and Cognitive Presumptions

Nevertheless, the proposal above is only provisional. It raises at least three important concerns. First, in some cases, p can be a practical presumption even when proceeding on non-p promotes the same goal and, thereby, the goal of equal normative importance (see Koplin & Selgelid 2015, pp. 601-602). Second, whether acquiring false propositions has the same cognitive cost as rejecting true propositions may depend on whether we take into account the logical consequences of a proposition. Imagine, for instance, that p has high inferential potential and leads to accurate predictions (even if p is false). Under this condition, the epistemic costs do not seem equal: acquiring false proposition p, all things considered, seems less costly than rejecting true proposition p. Third, it must be noted that some epistemological approaches allow goals of unequal normative importance. For instance, in virtue epistemology (see Turri et al. 2019), one may prefer intellectual humility over intellectual autonomy and, based on the virtues of unequal importance, estimate the cost of actions and proceed on a less costly proposition. These concerns are both important and plausible, but they deserve a separate treatment. Their purpose in this chapter is to explain the nature of the claim that the estimation of costs might be different in practical and cognitive cases. At this point, the claim is merely a hypothesis, but one worth considering and surely requiring further examination.

2

To some scholars, the difference between foundations of practical and cognitive presumptions might seem an obvious consequence that follows from the fundamental difference between the non-epistemic and epistemic contexts. The present analysis, however, assumed that the analogy between practical and epistemic contexts is worth exploring. Starting from this assumption, I sought to establish whether the pragmatic principles that, supposedly, underlie Rescher’s typical cognitive presumptions have a closely analogous rationale as the principles underlying paradigmatic practical presumptions. My conclusion is that the similarity with the pragmatic rationale is only nominal: even if avoiding greater harm underlies both practical and cognitive presumptions, the pragmatic policy comes down to avoiding costly errors only in the case of a paradigmatic practical presumption. As a result, practical and cognitive presumptions, strictly speaking, do not share the same pragmatic function. The differences are even greater. Not only do practical and cognitive presumptions have different dialogical (Section 3.2) and pragmatic functions (Section 3.3) but, sometimes, they cannot be defeated by the same types of evidence. Sometimes, as far as ‘all else being equal’ is concerned, practical and cognitive presumptions are not equal. In Section 3.4, I illustrate this difference.

77

CHAPTER 2

3.4 What Can Defeat Cognitive Presumptions, and When? To provide a fully-fledged account of the defeating conditions, a theorist must answer at least two questions: “What kinds of considerations, in principle, defeat presumptions?” and “When do these considerations, exactly, defeat presumptions?” As far as the first question is concerned, cognitive and practical presumptions are similar: they can be defeated by the tenability criticism, as well as by undermining, undercutting, overriding, and rebutting defeaters (see Section 2.3).77 With regard to the second question, however, cognitive and practical presumptions are different: sometimes, under the same background conditions, they cannot be defeated by the same types of evidence. For illustration, let us focus on the Cat case and the role of an undercutting defeater. Imagine that, after a general sceptical remark, Jane says: “Andy, the neighbour has a cat that looks exactly like yours” (u). Although consistent with both “Andy sees that the cat, which looks exactly like his cat, is in the tree” (q) and “Andy’s cat is in the tree” (p), this information reduces the reliability of the connection between q and p and seems to cancel Andy’s cognitive presumption. Nevertheless, I believe that cancelling p’s presumptive status depends on p’s prior likelihood and that, sometimes, u defeats a cognitive presumption p under different conditions than a practical presumption p. Suppose that before Jane puts forward an undercutting defeater, Andy and Jane attribute equal likelihoods to “Andy’s cat is in the tree” and “Andy’s cat is not in the tree.” Under this condition, both visual perception and prior likelihoods suggest that Andy should suspend judgment. On the one hand, visual impression is not a reliable indication that Andy’s cat is in the tree (since the neighbour’s cat looks exactly the same). On the other hand, prior likelihood is not a reliable indication that Andy’s cat is in the tree (since there is an equal chance that Andy’s cat is somewhere else). Consequently, “Andy’s cat is in the tree” and “Andy’s cat is not in the tree” seem like equally plausible propositions, and the original presumptive status must be cancelled.78 So, if prior likelihoods of p and non-p are equal (or unknown), and p is a cognitive presumption, then the undercutting defeater u cancels p’s presumptive status. Recall that, if p were a practical presumption, and conditions were the same, u would not cancel p’s presumptive status (see Section 2.3). Next, suppose that the prior likelihood of “Andy’s cat is in the tree” is sufficiently higher than the prior likelihood of “Andy’s cat is not in the tree” because, on a normal day, Andy’s cat spends most of the time in that tree. As far as his visual impression alone is concerned, Andy, of course, should suspend his judgment; but as far as prior likelihoods are concerned, he should proceed as if his cat is, indeed, in the tree. Thus, when all 77 

With the proviso that overriding in the cognitive case might be difficult to achieve under the assumption of the

normative equality of epistemic goals (acquiring truths vs. avoiding falsities). 78 

The additional assumption in all the variations of the Cat case is that Andy and Jane cannot estimate the likelihood

of “The neighbour’s cat is in the tree.” While they know that the neighbour’s cat looks very similar to Andy’s cat, they do not know anything about the cat’s behavioural patterns.

78

On the Differences Between Practical and Cognitive Presumptions

available evidence is taken into account, “Andy’s cat is in the tree” is the most plausible proposition and its presumptive status remains in place. Hence, if the prior likelihood of p is sufficiently higher than the prior likelihood of non-p, and p is a cognitive presumption, then the undercutting defeater u does not cancel p’s presumptive status. Recall that, if p were a practical presumption, and conditions were the same, u would indeed cancel p’s presumptive status (see Section 2.3). Finally, suppose that the prior likelihood of “Andy’s cat is in the tree” is sufficiently lower than the prior likelihood of “Andy’s cat is not in the tree” because, on a normal day, Andy’s cat spends only a couple of minutes in that tree. In these circumstances, the information that the neighbour’s cat looks exactly the same weakens the evidential strength of Andy’s visual impression, and forces him, first, to suspend judgment on “Andy’s cat is in the tree.” The prior likelihood, then, makes the suspension of belief implausible and instructs Andy to take a further step—to rebut “Andy’s cat is in the tree” and proceed on “Andy’s cat is not in the tree.” This leads us to the following conclusion: if the prior likelihood of p is sufficiently lower than the prior likelihood of non-p, and p is a cognitive presumption, then the undercutting defeater u cancels p’s presumptive status. Only under these conditions do undercutting defeaters operate analogously in the cognitive and practical cases.

2

To sum up, practical and cognitive presumptions merit distinct treatment in argumentation theory. Not only do they require different accounts of dialogical and pragmatic functions, but of defeating conditions as well. Although, they can, in principle, be defeated by the same kinds of considerations, sometimes, under the same background (dialogical, epistemic) conditions, they cannot be defeated by the same types of evidence. To support this point, I investigated the effectiveness of the undercutting defeater. This type of evidence will readily defeat a practical presumption p, but will not defeat a cognitive presumption p if the prior likelihood of p is sufficiently higher than the prior likelihood of non-p. Moreover, the undercutting defeater will not defeat a practical presumption p, but will readily defeat a cognitive presumption p if the prior likelihoods of p and non-p are equal (or unknown). The potential theoretical consequences of this difference are considerable and weighty. For instance, by showing that different background conditions can produce structural differences in the Challenger’s burden of proof, our analysis can have relevant deontic implications. Although presumptions, due to their deontic function, impose the obligation to present reasons on the Challenger, whether the Challenger should be obliged, for example, to undercut a presumption, could depend on whether she can indeed do so. Accordingly, our analysis might contribute to the investigation of how presumptions relate to burdens of proof. It might show that, if the Challenger bears the burden of proof, she might be expected to undercut a presumption p only if (1) p is a practical/cognitive presumption, and (2) the prior likelihood of p is (sufficiently) higher/equal/lower than the prior likelihood of non-p. Since the burden of proof is considered to be a central, defining notion in standard theories of presumption, this contribution should be worthy of attention. 79

CHAPTER 2

4 Conclusion For the most part, this chapter sought to achieve a typological objective. It explored various grounds for distinguishing practical and cognitive presumptions. In the literature, scholars typically distinguish the two types of presumptions by referring to what I call ultimate contextual functions: while practical presumptions facilitate the process of decision-making, cognitive presumptions facilitate the process of epistemic inquiry (see Ullmann-Margalit 1983, p. 143; Rescher 2006, p. 27). Also, scholars recognize that practical and cognitive presumptions have qualitatively different foundations (Godden & Walton 2007, p. 315, 338), or what I call distinct normative functions: while practical presumptions promote non-epistemic goals in deliberation, cognitive presumptions promote epistemic goals by introducing information to the inquiry. But what about other features and functions?79 Are practical and cognitive presumptions both conclusions of presumptive reasoning (logical function)? Do they asymmetrically allocate the burden of proof (deontic function)? Are they meant to overcome a dialogical impasse, and, if they are, are they dealing with a similar kind of impasse (dialogical function)? Are practical and cognitive presumptions both concerned with avoiding greater harm and, if they are, does this necessarily imply avoiding costly errors (pragmatic function)? Finally, given their equivalent, or closely similar background conditions, can they be defeated by the same kinds of evidence (defeating conditions)? The goal of this chapter was to explore the relationship between practical and cognitive presumptions in terms of dialogical functions, pragmatic functions, and defeating conditions. To achieve this typological objective, it was necessary to develop dialogical accounts of practical and cognitive presumptions, which constituted the dialectical objective. The starting point of my dialectical and typological research was Nicholas Rescher’s view that practical and cognitive presumptions are closely analogous. What justifies this starting point is surely the general relevance of Rescher’s extensive and influential work, but also a distinct theoretical potential of the strong analogy. Namely, if the strong analogy is plausible, then, first, the term ‘presumption’ preserves conceptual cohesion, and, second, the study of cognitive presumptions can directly benefit from the study of practical presumptions (and vice versa). However, I showed that practical and cognitive presumptions differ in several important respects. The main dialectical and typological results are summarized as follows.

79  The following questions might seem relevant to the issue concerning differences between legally inspired (practical) presumptions and ordinary, non-legal presumptions (see Kauffeld 2003). However, it is debatable whether Rescher’s cognitive presumptions really belong to the class of ordinary, everyday presumptions. Rescher’s cognitive presumptions might be applicable to everyday contexts to some extent, but they belong, first and foremost, to ideal dialectical settings and are concerned neither with social conventions nor with the rhetorical management of probative obligations incurred by the performance of speech acts (for details, see Kauffeld 1998, 2005).

80

On the Differences Between Practical and Cognitive Presumptions

Result 1. Typically, the two types of presumptions have different dialogical functions. On the one hand, in order to stop the (infinite) dialectical regress, cognitive presumptions are meant to block a particular type of dialogue. Also, their function is to enable the dialogue to get off the ground. On the other hand, in order to overcome evidential uncertainty in deliberation, practical presumptions unblock the dialogue. Their function is to allow the dialogue to proceed towards a desirable conclusion. Result 2. Typically, the two types of presumptions involve structurally different foundations. Unlike the scheme for presumptive practical reasoning, the complete scheme of presumptive cognitive reasoning does not seem to include premises related to making decisions under genuine uncertainty. This entails that cognitive presumptions do not arise in response to the (comparable) deliberation problem, do not serve to avoid costly errors, and, thus, have a different pragmatic function.

2

Result 3. Sometimes, under the same background conditions, the two types of presumptions cannot be defeated by the same types of defeaters. For example, sometimes, given a particular attribution of prior likelihoods, an undercutting defeater will cancel a practical presumption but will not cancel a cognitive presumption (and vice versa). As a result, the two classes of presumptions sometimes burden Challengers with structurally different dialectical obligations. The proposed conclusions are qualified on purpose. I do not wish to argue that all practical presumptions are different from all cognitive presumptions in every respect specified above.80 Rather, my overall, modest conclusion is the following: typical cognitive presumptions are usually different from paradigmatic practical presumptions in many relevant respects. Also, the proposed conclusions are only preliminary. In order to decisively separate cognitive from practical presumptions, additional theoretical effort is required, and a more sophisticated account of practical and cognitive presumptions is needed. This is primarily true, I believe, for the account of presumptive reasoning. For instance, our presentation of the foundations of presumptions discusses neither the relevance (or necessity) of the normative consensus between the dialogical parties81 nor the nature of the relation between (prudential) goals.82 However, I believe that the proposed conclusions, although preliminary, are at least strong enough to direct research towards less explored territories. At least, they are strong enough to show that practical and cognitive presumptions merit distinct dialectical treatment until proven otherwise. 80  On the one hand, Rescher’s (non-typical) cognitive presumption based on the principle of simplicity might well include a premise concerning genuine evidential uncertainty. On the other hand, legal scholarship acknowledges (practical) presumptions that are not genuinely uncertain as far as prior likelihoods and base rates are concerned: for instance, there is a presumption that a properly posted letter is successfully delivered to the recipient (see Gama 2017, p. 563; Petroski 2008, p. 390). 81  For the argument that normative consensus is necessary for the correct application of (practical) presumptions in social controversies, see Räikkä (1997, pp. 468-469). For a similar argument applied in the context of contemporary bioethics, see Koplin & Selgelid (2015, p. 601). 82  What justifies the contention that two ultimate goals are of equal or unequal importance? Are we talking about an ‘objective’ relation, or an interpersonally valid relation (agreed by parties in that particular dialogue), or, perhaps, a relation that is personally preferred by an individual agent?

81

CHAPTER 2

The latter typological and dialogical results open a straightforward conceptual question: what, in the end, holds practical and cognitive presumptions together? They indeed give rise to different types of status, but can they still be seen as different types of presumptions? In part, the answer depends on the preferred level of conceptual generality. On the one hand, one may contend that, despite the differences, both types of presumptions share a similar dialogical function—to enable the ‘constructive development of the dialogue’ (either by enabling progress or preventing regress). Or, one may contend that both types of presumptions share the same pragmatic function of ‘avoiding greater harm,’ even though this policy is not always selected on the grounds of avoiding probable error. On the other hand, even if one prefers a lesser degree of conceptual abstraction (as I personally do), the two types of presumptions may still share the same deontic function. To see whether this is indeed true, one would need to develop a clear conception of the burden of proof and explore whether this conception equally applies to practical and cognitive presumptions. I deal with this complicated task in Chapter 3 (i.e., Bodlović 2020b). To conclude, one may still find it conceptually plausible to separate practical and cognitive presumptions from many other phenomena, such as presuppositions, suppositions, assumptions, stipulations, assertions, claims, or hypotheses (see Godden 2017). Although this chapter discusses the ambiguities of some fundamental concepts (e.g., dialogical progress, pressure, uncertainty, avoiding harm, etc.) and thereby contributes to the underlying conceptual objective, it is primarily concerned with the dialogical application of presumptions. To use a metaphor, the proponent who would use the ‘Practical Presumption Instruction Manual’ on a cognitive presumption might get into all sorts of trouble, regardless of what she believes about the conceptual connection between the two. Due to Result 1, the User might forget what she was trying to achieve in the first place; due to Result 2, the User might misjudge her reasoning options and, due to Result 3, get the wrong picture about when her presumption becomes defeated. This chapter aims to get this imaginary User out of trouble—to provide some preliminary instructions on how to deal with practical and cognitive presumptions in ordinary dialogues. Acknowledgments: The earlier version of this chapter was presented at the conference Nove teme u filozofiji (Zagreb, November 29-30 2018), as well as at the 3rd European Conference on Argumentation Reason to Dissent (Groningen, June 24-27 2019). I thank the organizers of these conferences, as well as the participants for their valuable comments and suggestions. I am especially indebted to my supervisors dr. Jan Albert van Laar and dr. Barteld Kooi, my ECA commentator dr. James Freeman, two reviewers and editors of Argumentation, Viktor Ivanković, Karolina Kudlek, and Diego Castro Amenábar. Finally, I thank the chair of the committee, dr. Christopher Tindale, and the members of the committee for awarding the earlier version of this chapter with the ‘Frans van Eemeren Prize’ for an outstanding student paper.

82

On the Differences Between Practical and Cognitive Presumptions

2

83

3

CHAPTER 3 On Presumptions, Burdens of Proof, and Explanations

CHAPTER 3

My task is to explain to you as quickly as possible my essence, that is, what sort of man I am, what I believe in, and what I hope for, is that right? —Dostoevsky, The Brothers Karamazov

I went close up to him and made a last attempt to explain that I’d very little time left, and I wasn’t going to waste it on God. —Camus, The Stranger

86

On Presumptions, Burdens of Proof, and Explanations

On Presumptions, Burdens of Proof, and Explanations Abstract: On the standard view, all presumptions share the same deontic function: they asymmetrically allocate the burden of proof. But what exactly does this function amount to? Once presumptions are rejected, do they impose the burden of arguing, the burden of explanation, or the most general burden of reasoning on their challengers? In this chapter, I take into account the differences between cognitive and practical presumptions and argue that the standard accounts of deontic function are at least ambiguous (because the two types of presumptions entail distinct conceptions of the ‘burden of proof’), and likely implausible. As a result, they require qualifications. Keywords: adequacy, asymmetrical allocation, burden of explanation, burden of proof, cognitive presumption, deontic function, Flat Earth, practical presumption.

3

1 Introduction83 The notion of burden of proof plays an important role in both ordinary and academic parlance. Broadly understood, it denotes the dialogical obligation of a party (e.g., proponent) to support her position once it gets challenged, rejected, or contradicted by the opponent. In law, philosophy, and argumentation theory, the burden of proof raises many questions, but one stands out in terms of popularity and practical relevance: Which party in the discussion carries, or should carry the burden of proof? How should we, in light of this, structure a reasonable discussion? If both parties have standpoints, should they both carry an (equal) burden of proof, or should one party be dialectically privileged? I shall call this the allocation question. In this chapter, I deal with the allocation question in situations where a presumption is challenged, rejected, or contradicted. In standard cases, where it is not the presumption that gets challenged, it might seem that there is a simple answer: in the mixed discussion, where both parties have standpoints, we should accept a symmetrical allocation of the burden of proof (van Eemeren & Houtlosser 2002; Walton 2014). This answer coheres well with so-called dialectical egalitarianism. According to this philosophical tradition, if the parties have reached a prior decision to engage in a reasonable discussion, every proposition, when challenged, must be supported by an argument (see Rescorla 2009a). Consequently, neither party or standpoint is privileged: “In a mixed dispute, where two parties have advanced contradictory standpoints, each party has a burden of proof for his own standpoint” (van Eemeren & Houtlosser 2002, p. 18).

83  This chapter is a revised and extended version of Bodlović (2020a).

87

CHAPTER 3

Despite their intuitiveness and elegance,84 symmetrical allocation and dialectical egalitarianism are, to some extent, controversial. Many philosophers and argumentation scholars argue that we should sometimes accept an asymmetrical allocation of the burden of proof, given that some propositions are dialectically privileged. Once these “onusassigning statements” (Brown 1970, p. 74) get rejected, they do not require support until or unless the Challenger puts forward (sufficient) reasons against their acceptability. According to the standard account, privileged propositions, which shift, reverse, or asymmetrically allocate the burden of proof, and that are defeasible in nature, are called presumptions (Whately 1963; Rescher 1977, 2006; Ullmann-Margalit 1983; Cargile 1997; Freeman 2005; Walton 2014; Godden 2017). Thus, if both parties have standpoints and their standpoints are challenged, then both parties carry the burden of proof unless one standpoint is a presumption. Standard accounts of presumption are surprisingly heterogeneous (see Godden & Walton 2007; Lewiński 2017). There are at least two sources of heterogeneity. First, there are significant disagreements among scholars on how to characterise presumptions,85 and, second, there are fundamental differences between different types of presumptions. For instance, we can distinguish between cognitive (epistemic) and practical presumptions, since they operate in different dialogical contexts (epistemic inquiry vs. practical deliberation), perform distinct normative functions (promoting epistemic vs. non-epistemic goals), and thereby exhibit qualitatively different foundations (Ullmann-Margalit 1983, p. 143; Rescher 2006, p. 27; Godden & Walton 2007, p. 337; Lewiński 2017, p. 610).86 Cognitive presumptions are typically used in the context of inquiry. In a reasonable epistemic discussion about the shape of the Earth, “The Earth is globe-shaped” is a (strong) cognitive presumption since this proposition “represents our most plausible candidate for truth” (Rescher 2006, p. 71). The dialogical parties should recognize that, due to overwhelming and well-known (scientific) evidence that the Earth is globe-shaped, “The Earth is globe-shaped” should be exempted from the standard burden of proof, i.e., the burden of proof should rest 84  One aspect of ‘elegance’ is that it almost seems possible to resolve the allocation question on conceptual grounds. That is, if we must attribute the burden of proof to every party whose position gets challenged (rejected, contradicted) in the dialogue, then allocating the burden of proof comes down to nothing more than rephrasing the definition of the burden of proof: the answer to the question of who carries the burden of proof seems already contained in the meaning of the burden of proof. 85 

For instance, scholars disagree whether presumptions are rules (Prakken & Sartor 2006) or propositions (Walton

2014); statuses, modifiers, and qualifiers of propositions (Godden 2017, 2019) or relations (Pinto, according to Freeman 2005, p. 27); based on inferential (Godden & Walton 2007) or discursive support (Freeman 2005; Rescher 2006); and whether they belong to the opening (van Eemeren & Houtlosser 2002) or the argumentation stage of the critical discussion (Walton 2014). 86 

Importantly, the proposed distinction is also a matter of controversy. Some scholars believe that we should indeed

distinguish practical from cognitive presumptions (e.g., Rescher 2006, Freeman 2005), and others are sympathetic to this possibility (e.g., Ullmann-Margalit 1983, p. 143; Godden & Walton 2007, p. 337). However, Godden (2019) has recently argued against the theoretical relevance of cognitive (epistemic) presumptions and insisted that ‘presumably’ is exclusively a practical (non-epistemic) modality. Although he presents plausible concerns about the notion of cognitive presumption, this chapter sets these concerns aside and assumes that cognitive presumptions represent a legitimate class of presumptions.

88

On Presumptions, Burdens of Proof, and Explanations

only with its Challenger. The concept of cognitive presumption coheres well with dialectical foundationalism, a philosophical tradition suggesting that the rules of reasonable discussion should protect (epistemically) uncontroversial propositions and sanction (epistemically) deviant challenges (Rescorla 2009a). By contrast, practical presumptions are typically used in contexts of practical deliberation. In criminal law, for instance, “John is innocent” is a (strong) practical presumption: since we should “try to minimize the conviction of innocent persons, even at the cost of letting guilty persons go free” (Walton 1988, p. 244) the (global) burden of proof rests only with the prosecutor, meaning that “John is innocent” stands until or unless the prosecutor proves otherwise. Although inspired by legal scholarship, the concept of practical presumption is readily applicable to many everyday contexts where, in circumstances of uncertainty and pressure to make a timely decision, we try to avoid more significant harm (see Ullmann-Margalit 1983; Godden 2017). Cognitive and practical presumptions are different in many respects,87 but one feature seems to hold them together. According to standard accounts, regardless of whether p is a cognitive or a practical presumption, if p gets rejected, then the burden of proof rests exclusively with the Challenger. Hence, all presumptions share the same deontic function: they asymmetrically allocate the burden of proof. Asymmetrical allocation is a cornerstone of the standard account of presumptions for two reasons: first, all types of presumptions, supposedly, share this function (despite other differences), and, second, all scholars seem to agree on this (despite having other disagreements).

3

In this chapter, I will explore the deontic function of presumptions in some detail and (provisionally) answer the allocation question insofar as presumptions are concerned. The main question is the following: What exactly does the deontic function of presumption amount to? Once presumptions are rejected, is the burden of proof immediately shifted onto the challengers? If it is, what exactly does the burden of proof amount to; if it is not, what other obligations might be relevant? Do presumptions impose the burden of arguing, the burden of explanation, or the most general burden of reasoning on their challengers? And in what way are obligations to provide arguments, explanations, or reasons (of whatever kind) connected to the burden of proof? At the very least, the chapter shows that the standard account of the deontic function is ambiguous. If presumptions allocate the burden of proof to the Challenger, then they allocate it in two distinct senses: cognitive presumptions allocate a general burden of reasoning (challengers are obliged to produce any kind of reason), and practical presumptions allocate a more specific burden of arguing (challengers are obliged to produce dialectically adequate reasons). Nevertheless, the standard account is not only ambiguous but also implausible: 87 

Not only do cognitive and practical presumptions have different ultimate contextual functions (i.e., acquiring truths

vs. making right decisions) and different normative functions (i.e., promoting epistemic vs. non-epistemic goals), but they also have distinct dialogical functions (i.e., practical presumptions do not stop dialectical regress), pragmatic functions (i.e., the two types of presumptions usually do not avoid greater harm in the same sense); and, at least sometimes, they seem susceptible to different types of defeaters. For details, see Bodlović (2021).

89

CHAPTER 3

since the burden of proof is best understood as the burden of arguing (rather than the burden of reasoning, which includes the burden of explanation) cognitive presumptions, strictly speaking, do not asymmetrically allocate the burden of proof. Consequently, the standard account requires clarifications and, most probably, revisions. To be sure, some scholars have already claimed that the standard account of presumption, as well as dialectical foundationalism, are implausible. For instance, Kauffeld (1998, 2003, 2005) and Bermejo-Luque (2019) argued that, in the context of everyday dialogue, presumptions are not always negatively correlated to the burden of proof, and dialectical egalitarians insisted that a reasonable discussion should never be dogmatic (Rescorla 2009a, p. 96). This chapter reinforces this trend by distinguishing between various types of presumptions and distinct possible meanings of the burden of proof, and answers the more refined allocation questions generated by a more detailed typological and conceptual analysis. For the most part, the chapter hopes to advance the study of presumptions by adding to its analytic sophistication. I first outline the common principles of asymmetrical allocation (see Section 2), and then focus on cognitive and practical presumptions (see Section 3). In Section 4, I deal with the different possible meanings of the burden of proof (the burden of reasoning, and the burden of arguing) and their relationships to the burden of explanation. After spelling out these notions, I explain how they relate to cognitive and practical presumptions (see Section 4.1—4.4). In conclusion, I provide a summary of the most important results.

2 Common Principles of Asymmetrical Allocation The idea that, sometimes, only one dialogical party must provide reasons when parties have different standpoints is present in both ordinary and academic contexts. For instance, when a journalist challenged Donald Trump’s suspicions about potential voter fraud, Trump refused to support his suspicions. Instead, he complained that the journalist did not provide evidence that fraud did not happen. Trump thus implicitly appealed to the asymmetrical allocation of the burden of proof (for details, see CNBC.com staff 2016). In academic contexts, appeals to asymmetrical allocation are often more explicit. It is “a piece of scientific orthodoxy” (Dare & Kingsbury 2008, p. 511) that the burden of proof should rest exclusively with those who propose new scientific hypotheses, theories, or paradigms. Asymmetrical allocation is also advocated by philosophers. In a metaphysical discussion on whether free will and determinism are compatible, Lycan (2003) argues that only the proponents of incompatibilism should carry the burden of proof. In the discussion about the existence of God, Pascal (2004) suggests that the burden of proof should rest with the atheist (‘Pascal’s wager’), whereas Scriven (1966) and Flew (1972) argue that a presumption should favour the atheist. In epistemology, an asymmetrical allocation regulates the debate between the sceptic and the dogmatist (Brown 1970; Kelly 2005), and resolves the problem of infinite (dialectical) regress (see Rescher 1977; Rescorla 90

On Presumptions, Burdens of Proof, and Explanations

2009b). Asymmetrical allocation is also present in ethical debates on discrimination and global justice (Räikkä 1997; 2005), organ markets, human enhancement, climate change, and the precautionary principle (Koplin & Selgelid 2015). Finally, in metaphilosophy, Williamson (2011) argues that the burden of proof should rest with those who deny that philosophical expertise contributes to the successful performance of thought experiments. Therefore, asymmetrical allocation has been advocated by a number of scholars, across many disciplines and within different intellectual traditions. Asymmetrical allocation of the burden of proof can be “merely a labor-saving device” (Brown 1970, p. 74), a trick to gain an unfair dialogical advantage (as in the Trump example). Yet, asymmetrical allocations can have procedural, prudential, epistemic, and even moral foundations that generate various principles. In my view, we can divide the common principles of asymmetrical allocation (which are not motivated exclusively by rhetorical self-interest) into two broad categories: proposition-related principles and agent-related principles. In the remainder of this section, I present some well-known principles and explain the proposed categorization.

3

Consider the popular slogan ‘she who asserts must prove.’ Although this “overarching principle of burden of proof” (Walton 2014, p. 99) appears straightforward, it is ambiguous for (at least) two reasons. First, it is unclear whether ‘she who asserts must prove’ is a principle of symmetrical or asymmetrical allocation. On the one hand, if we understand the principle literally and test it in a mixed discussion, then the principle allocates the burden of proof symmetrically: both parties are obliged to provide reasons because both parties assert standpoints. On the other hand, it seems that ‘she who asserts must prove’ is, at least sometimes, understood as a principle of asymmetrical allocation. The literature exhibits several (incompatible) explanations for why this might be the case, which, in turn, renders the overarching principle even more ambiguous. As a result, not only is it unclear whether the overarching principle allocates the burden of proof symmetrically or asymmetrically, but also on what exact grounds is this principle supposed to allocate the burden of proof asymmetrically. There are (at least) three interpretations of ‘she who asserts must prove’ that would allow the overarching principle to allocate the burden of proof asymmetrically in a mixed discussion. First, the act of making an assertion is sometimes understood as the act of ‘bringing the action’ or initiating an argument. On this interpretation, the overarching principle states that (1) the burden of proof rests (only) with the dialogical party who presents the view first, rather than with anyone who makes an assertion: “Whichever side initiates the assertion of a thesis within the dialectical situation has the burden of supporting it in argument” (Rescher 2006, p. 15).88 Second, the act of making an assertion is sometimes understood as the act of challenging accepted practices, values, or beliefs, i.e., the act of rejecting the status quo. On this interpretation, the overarching principle states that (2) the burden of proof rests (only) with the party who asserts something 88  This evokes the legal principle: Semper necessitas probandi incumbit ei qui agit.

91

CHAPTER 3

(dialectically, epistemically, or instrumentally) controversial (see Räikkä 1997, pp. 466467).89 Third, it seems that asserting is sometimes understood as ‘affirming.’ On this interpretation, the overarching principle states that (3) the burden of proof rests (only) with the party who asserts a positive claim (Cargile 1997, p. 61).90 For the most part, these three interpretations are concerned with the question: How should we allocate the burden of proof given what we know about the (normatively) relevant properties of standpoints and claims (i.e., is the claim presented first, is it controversial, and is it an affirmation or a negation)? As a result, the three interpretations of the overarching principle can be jointly labelled the proposition-based principles of asymmetrical allocation.91 The second group of principles concerns a different question, namely: How should we allocate the burden of proof (in order to facilitate optimal resolution) given what we know about the dialectically relevant abilities and inclinations of dialogical parties? Consider the case where David tells Luke: “Well, I paid for the last round of drinks, so it’s your turn!” Since it is in fact Luke who paid for the last round, he disagrees with David, and the discussion begins. But suppose that Luke holds a receipt in his pocket. In this case, although David initiates the conversation and advances a controversial claim,92 the resolution procedure will be more effective if Luke incurs the burden of proof. Luke has access to decisive evidence, and if he were obliged to present it, the issue would be quickly resolved. The principle, therefore, is the following: (4) The burden of proof rests (only) with the party who has better epistemic access to evidence.93 Finally, suppose that David wants to buy a car from Luke. In this case, it is crucial to determine whether the vehicle is in good condition, and Luke, the long-time owner, surely has better access to evidence. However, since Luke’s motivation to speak truthfully might 89 

Since parties can initiate a discussion without asserting a highly controversial claim, these principles are logically

independent. 90 

The principle “affirmati non neganti incumbit probatio” rests on the idea that it is not dialectically reasonable to

allocate the burden of proof to the proponent of the negative claim because it is demanding to prove the absence of property, fact, or incident. The standard phrasing, however, is even stronger, suggesting that it is impossible to prove a negative claim: “You cannot prove a negative.” This slogan is typically applied in discussions about the existence of paranormal or supernatural entities (such as God) and is “combined with the thought that people should not be assigned obligations to do impossible things” (Cargile 1997, p. 61). Nevertheless, scholars have argued that, in standard circumstances, it is possible to prove a negative claim (Cargile 1997; Macagno & Walton 2011; Pigliucci & Boudry 2014) and that even if proving negative claim is more demanding, it should not affect how the burden of proof is allocated. For instance, according to the “priority principle,” the negative assertion should only affect the order of defence: both parties should carry the burden of proof, but the proponent of the positive claim should go first (van Eemeren & Houtlosser 2002). 91 

One could argue that the previous allocation principles do not represent different interpretations of ‘she who asserts

must prove.’ Instead, they represent independent principles showing that the overarching principle is defeasible. On this view, ‘she who asserts must prove’ is a default principle that can be overridden by more specific considerations: the dialogical party should support her assertion unless (1) she did not initiate the discussion, (2) her assertion is uncontroversial, or (3) she asserted a negation. This alternative proposal would render the overarching principle less ambiguous. 92 

Suppose, for instance, that all of their other friends saw who paid for the drinks.

93 

In civil law, this is known as the “principle of fairness” (van Eemeren et al. 2002, p. 115) and is usually applied in

cases of product liability (Hahn & Oaksford 2007, p. 41).

92

On Presumptions, Burdens of Proof, and Explanations

be compromised, it is useful to allocate the burden of proof to David, who has a genuine interest in discovering the truth. If the primary goal is to acquire true beliefs about the car’s condition, then asymmetrical allocation might, once again, facilitate optimal resolution (see Dare & Kingsbury 2008).94 Hence, the underlying principle is the following: (5) The burden of proof rests (only) with the party who is less biased, or whose motivation is not corrupted. Notice that the last two principles do not concern the properties of propositions but rather the distinctive abilities or inclinations of dialogical parties (the ability to provide sufficient proof, and the inclination towards truth). They represent agentbased principles of asymmetrical allocation. In the following sections, I will mostly focus on the second proposition-based principle mentioned above. According to this principle, some propositions do not require support because, in specific contexts and circumstances, they seem epistemically plausible or instrumentally desirable. Typically, these propositions are called cognitive and practical presumptions. Although one may suggest that all of the aforementioned principles “explicitly or implicitly determine, at least partly, what it is reasonable or justified to presume” (Räikkä 1997 p. 463), the analysis in this chapter will start from a somewhat narrower concept of presumption.

3

3 Cognitive and Practical Presumptions Suppose that Diane and Steve are discussing the shape of the Earth. Diane admits that scientific claims are defeasible but believes that science offers the best available answers about the natural world. In this particular epistemic dialogue, Diane claims that the Earth is globe-shaped (spherical) and readily accepts this belief as her standpoint. By contrast, Steve is a member of the Flat Earth Society, and believes that scientific evidence is a fabrication. He thinks that the Earth is disc-shaped, and accepts this belief as his standpoint. In the ordinary context, “The Earth is globe-shaped” is a (strong) cognitive presumption. Since this empirical claim is, in principle, defeasible, it could be reasonable to discuss it and, should contradictory evidence arise, even reject it.95 In normal cases, 94  This is linked to the legal principle of caveat emptor (Dare & Kingsbury 2008). The proposed list is not exhaustive. For instance, Hansen (2019) mentions two additional principles: the burden of proof “belongs to the party who would lose the case if the outcome had to be decided at this point,” and “it belongs where the law says it belongs” (pp. 15-16). In addition, we can refine the study of asymmetrical allocation by taking into account the pragmatic dimensions of propositions. In other words, we should not focus only on the party who introduces the proposition (e.g., is it the party who ‘brings the action,’ or with ‘pure motivation’?) and the nature of proposition (e.g., is it negative, or perhaps controversial?), but also on the (intended) pragmatic goal that underlies the proposition. Along these lines, Kauffeld (1998) suggests that allocating the burden of proof depends (1) on the type of speech act. Thus, p will entail the burden of proof if used as a proposal, but not if used as a suggestion. Pragma-dialecticians also recognize the importance of pragmatic properties. In their opinion, however, what determines the allocation is, strictly speaking, not the nature of the particular speech act, but (2) the argumentative function of this speech act in a critical discussion (Tseronis 2009, p. 83). Thus, some suggestion p will entail the burden of proof if it represents a (sub)standpoint, but not if it represents a starting point accepted at the opening stage. 95  “The Earth is globe-shaped” is defeasible in the sense that it cannot be claimed with mathematical certainty that the proposition is true, and it is conceptually possible that some day we will obtain information that would show it to be false. So, it is not, in principle, irrational either to have a discussion about it or to reject it.

93

CHAPTER 3

however, any reasonable person (with primary education) should nowadays acknowledge that, in Rescher’s words, “The Earth is globe-shaped” is “our most plausible candidate for truth” (2006, p. 71). Many reliable epistemic sources vouch for it,96 and it requires dialectical recognition: in epistemic dialogue, the proponent of a plausible standpoint and the proponent of an implausible standpoint should not play by the same rules. Granted, dialectical rules should not require the immediate acceptance of the most plausible standpoint, but neither should they force us to proceed as if, initially, all propositions are equally plausible. Epistemic dialogues must avoid both uncritical dogmatism and naïve egalitarianism: on the one hand, its rules must allow criticism of plausible standpoints (since this is a desirable epistemic attitude) but, on the other hand, they must give some dialectical credit to (highly) plausible standpoints (since, initially, they seem epistemically warranted). If we must regulate epistemic dialogues by dialectical rules that systematically further epistemic ends, then the asymmetrical allocation of the burden of proof seems like a proper regulatory choice. In the Flat Earth example, the acceptance of “The Earth is globe-shaped” appears to facilitate the acquisition of true or justified beliefs, so Diane should not carry the burden of proof. By contrast, the initial acceptance of “The Earth is disc-shaped” appears to get in the way of achieving epistemic goals, and, for this reason, Steve should carry the burden of proof.97 This means that Steve should show that circumstances are somehow exceptional, and present (persuasive) arguments that the Earth is disc-shaped.98 If his arguments ultimately fail to withstand Diane’s scrutiny, “The Earth is globe-shaped” is accepted by default.99 Thus, initially, asymmetrical allocation favours the more plausible propositions and sanctions deviant challenges, but still allows credible claims to be criticized and offers a normative infrastructure for having a debate about widely-accepted views that happen to be controverted by eccentric individuals. One may wonder whether a member of the Flat Earth Society is committed to the evidence that supposedly makes “The Earth is globe-shaped” a cognitive presumption. If Steve finds (scientific) evidence inadequate to begin with, then it is unclear whether “Steve 96 

For instance, scientific authority, evidence, testimonies, and explanatory utility.

97 

Not only does this view cohere with some pre-theoretical intuitions (Rescorla 2009a), but it also seeks to resolve

a well-known theoretical problem of infinite (dialectical) regress (see Rescher 1977; Freeman 2005; Rescorla 2009b). 98  Notice that Steve is obliged to present (persuasive) arguments that the Earth is disc-shaped in order to win the discussion, but not in order to defeat the presumptive status of “The Earth is globe-shaped.” The presumptive status of Diane’s standpoint gets defeated as soon as Diane incurs the regular burden of proof, i.e., becomes obliged to argue that the Earth is globe-shaped. In principle, Steve can achieve this by successfully undercutting well-known scientific evidence, thereby showing that “The Earth is disc-shaped” and “The Earth is globe-shaped” are equally plausible propositions. “The Earth is globe-shaped” would then cease to be the most plausible truth-candidate, but Steve would not win the discussion by defeating its presumptive status in the described fashion. Instead, he only resets the discussion: from that point on, Diane also incurs the burden of proof, and the dialogue continues in a new normative setting. 99 

This dialectical bias towards the scientific status quo is relevant for regulating the debate between scientists and

pseudoscientists. According to Pigliucci & Boudry (2014), pseudoscientific claims (concerning, e.g., Intelligent Design, extra-terrestrial visits, anti-vaccination, or the denial of an anthropogenic climate change) have low initial probabilities, so their proponents should carry the burden of proof.

94

On Presumptions, Burdens of Proof, and Explanations

actually rejects a presumption, rather than Diane’s assertion of the presumptive status” (Godden 2020, p. 5). This elicits a fundamental question: Can p have presumptive status in a dialogue D if reasons R that support p (i.e., make p the most plausible proposition) are not conceded by parties in D, i.e., are not dialectically adequate in D? In standard theories of cognitive presumption, the dialectical adequacy of R does not seem necessary for p’s presumptive status. There are at least two reasons for this. First, if the dialectical adequacy of R were necessary, then, at some point in the dialogue, the Challenger could nullify the presumptive status of p by making R inadequate: the Challenger could simply retract her commitment to (the previously conceded) reasons R and, by doing so, defeat p’s presumptive status. On standard accounts, however, this does not suffice: to defeat p’s presumptive status, the Challenger must offer reasons (and not only retract some). Second, the Challenger could be committed to a presumption because of the broader cultural context and, in some sense, intellectual humility. In other words, even if Steve neither concedes R nor recognizes its force, he should still recognize that, in the relevant cultural context, R and the presumptive status of “The Earth is globe-shaped” are accepted by the (more or less) equally competent majority. Thus, even if Steve rejects the adequacy of these reasons and doubts their plausibility, he must still recognize their legitimacy. The further investigation of the normative significance of plausibility, adequacy, and legitimacy, and their relationship, falls outside the scope of the present chapter.

3

To sum up, on the standard view, if p is the most plausible truth-candidate in inquiry, then p is a dialogically privileged proposition and represents a cognitive presumption. The status of cognitive presumption is thus both epistemic, as well as dialogical status. Next, suppose that Mark and Alice are discussing whether to attend their friend’s house party. They are invited and would like to go, but their decision (mostly) depends on whether John will be at the party. Mark and Alice despise John, and agree that his presence would surely ruin their evening.100 Seeing that the chance that John will be at the party is roughly 50%, Mark and Alice are facing the so-called “deliberation problem” (UllmannMargalit 1983, p. 152; Godden 2017, p. 505): since the party is about to begin, they must make their decision quickly, but John’s whereabouts are uncertain, and they cannot base their decision on a justified belief. Let us suppose that, in the described circumstances, Mark decides to proceed as if John will be at the party, and makes the following proposal: “We should skip the party.” Alice, however, replies: “We should go to the party.” In the Party example, “John will be at the party” is a practical presumption. Scholars typically characterise practical presumptions as propositions that we proceed upon (or take as true) when we are pressured to make a timely decision and the evidence is uncertain (Ullmann-Margalit 1983; Walton 2014; Godden 2017). In these special circumstances, the main goal is to make a decision ensuring the avoidance of more significant harm. 100 

For the sake of example, imagine that Mark and Alice have objective reasons to hate John, and that, due to these

reasons, John is the last person they would like to see, let alone spend time with him at the party. For instance, just a few months ago, John has seriously injured their daughter in a car accident while driving drunk and has never apologized.

95

CHAPTER 3

Accordingly, “John will be at the party” is a practical presumption because (1) there is pressure to decide whether to attend the party; (2) this decision depends on whether John will be at the party; (3) it is uncertain whether John will be at the party; but (4) parties agree that proceeding on “John will be at the party” and skipping it (when, in fact, John will not be there) seems like a lesser evil than proceeding on “John will not be at the party” and attending it (when, in fact, John will be there).101 Since proceeding on “John will be at the party” seems more desirable in the described circumstances to both Mark and Alice, the distribution of dialectical obligations should not be (completely) symmetrical. In a deliberative dialogue, the asymmetrical allocation of the burden of proof is a reasonable regulatory choice to avoid greater harm. Since proceeding on “John will be at the party” promotes this goal in the Party case, Mark should not carry the burden of proof: in circumstances of pressure and uncertainty, his position should be a dialogically privileged proposition. By contrast, proceeding on “John will not be at the party” potentially gets in the way of avoiding greater harm and, for this reason, Alice must carry the burden of proof. This means that Alice must provide arguments against the presumption, and if her arguments ultimately cannot persuade Mark to go to the party,102 “We should skip the party” is accepted by default. Dialogical parties should not carry equal dialogical obligations since, in the described circumstances, their practical standpoints are not equally desirable. In summary, if proceeding on p is the most desirable practical alternative in deliberation, then p is a dialogically privileged proposition and represents a practical presumption. Hence, the status of a practical presumption is both practical, as well as dialogical.103 In the next section, I explore the dialogical status by carefully examining the deontic functions of cognitive and practical presumptions.

4 Presumptions and Different Conceptions of the Burden of Proof The idea that presumption is “closely connected to the burden of proof” (Walton 2014, p. 117) lies at the heart of every standard account of presumption. According to Rescher, for instance, presumption and the burden of proof “represent correlative conceptions” (2006, p. 14) and are “opposite sides of the same coin” (2006, p. 14). Freeman seems to agree with this characterization since, according to his account, presumption is “the opposite of 101 

Notice that this everyday example is strikingly similar to the legal presumption of innocence. Here, “John is

innocent” has the status of practical presumption since (1) there is legal pressure to decide whether to convict John; (2) the decision depends on whether John is innocent or guilty; (3) it is evidentially uncertain whether John is innocent or guilty; but (4) acting as if John is innocent and setting him free (when, in fact, John is guilty) is treated as a lesser evil than acting as if John is guilty and convicting him (when, in fact, John is innocent). 102  She must prove that it is more likely that John will skip the party by using dialectically adequate reasons, i.e., reasons that Mark can be expected to accept. 103 

For a detailed characterization of practical presumptions, see Ullmann-Margalit (1983), Godden (2017), and

Bodlović (2021).

96

On Presumptions, Burdens of Proof, and Explanations

burden of proof” (2005, p. ix). But what, exactly, does this mean in a discussion where two parties take turns? Let’s begin with some clues from the literature. [A] ‘Presumption’ in favour of any supposition … implies that it must stand good till some sufficient reason is adduced against it; in short, that the Burden of proof lies on the side of him who would dispute it. (Whately 1963, p. 112) [W]here a proposition has the status of a presumption, then the burden of proof lies with anyone who refuses to concede it. (Pinto 2001, p. 4) [P]resumption is defined as a modal status (or property) of a claim (or proposition) indicating that the burden of proof with respect to that claim rests with anyone who would reject it. (Godden & Walton 2007, p. 315) The presumption shifts a burden [of proof] to the other side to disprove it, or the proposition becomes lodged into place as a commitment of both sides. (Walton 2014, p. 117)

3

[T]he presumption opponent is charged with something: with the burden of showing that not-q. (Ullmann-Margalit 1983, p. 151) Setting subtle differences aside,104 standard accounts all seem to accept the following: if User U puts forward p (that in the context at hand has the status of a presumption) at t1, and Challenger C rejects or challenges p at t2, then U does not carry the burden of proof at t3 whereas C incurs the burden of proof in (C’s next turn) t2+n. I shall call this the asymmetrical allocation of the burden of proof. The next question is: Is asymmetrical allocation unconditional, or is it conditional upon the User’s request for reasons? In other words, does Challenger C incur the burden of proof at t3, immediately after rejecting the presumption at t2? Or does C incur the burden at t4, after User U activates C’s burden by requesting reasons at t3? Does turn t2+n amount to t3 or t4? (t1) U: Presumably, p. (t2) C: Reject: “Presumably, p.” (t3) ? (t4) ? Although I do not discuss this issue in the present chapter, notice that the definitions in the literature say nothing about the active role of the User in allocating the burden of proof. C appears to incur the burden of proof by merely disputing, rejecting, or refusing to concede the presumption. This could indicate, at least to some degree, that on standard accounts 104 

That every presumption ‘implies’ or ‘indicates’ that the Challenger who rejects it carries the burden of proof does

not necessarily mean that every presumption ‘shifts’ the burden of proof.

97

CHAPTER 3

the asymmetrical allocation is supposed to be immediate: by rejecting the presumption at t2, C incurs the burden of proof at t3, and C’s burden is not conditional upon U’s request for reasons. I call this the unconditional allocation of the burden of proof and propose the following specification of the deontic function. The standard deontic function of presumption: Presumptions (1) allocate (2) the burden of proof in an (3) unconditional and (4) asymmetrical manner. To fully understand the standard deontic function of presumption, we should clarify four complicated notions: allocation, burden of proof, unconditional (allocation), and asymmetrical (allocation). In this chapter, I focus on the notion of the burden of proof and seek to provide a preliminary conceptual analysis. This is a surprisingly demanding task. Although we use the term casually in ordinary parlance, it is complicated to determine its precise meaning. In legal scholarship, for instance, Thayer claims that the burden of proof is an ambiguous notion that needs to be thoroughly explored and discriminated (2019, p. 75), and McCormick describes it as “the slipperiest member of the family of legal terms” (Strong 1992, p. 449). It is hardly surprising that as “the most successful jurisprudential ‘export’” (Hahn & Oaksford 2007, p. 40), the term retains its complexity in philosophy and argumentation theory. In the following sub-sections, I offer two interpretations of the burden of proof and analyse whether cognitive and practical presumptions, under these different interpretations, impose the burden of proof on the challengers.

4.1 Presumptions and the Burden of Reasoning Let’s begin with a basic underlying idea that everyone accepts. According to a “first approximation, having a burden of proof is being under an obligation … to support one’s view” (Aijaz et al. 2013, p. 260) or “furnish support for the thesis” (Johnson 2000, p. 194). Since the burden of proof naturally belongs to a context of dialogue (Rescher 2006, p. 19) and in this context ‘to support’ means to provide reasons, the burden of proof can be understood as an obligation to “provide reason” (Kauffeld 1998, p. 263) or “present some reasons” (Räikkä 1997, p. 467).105 I label this very general interpretation of the burden of proof the burden of reasoning (BoR): The burden of reasoning (BoR) represents the party’s dialogical obligation to provide any reason (either dialectically adequate or inadequate) for a position (view). What does this interpretation of the burden of proof mean for the deontic function of presumption? It means that, after rejecting a presumption, the Challenger is obliged to 105 

Though the basic idea is uncontroversial, to interpret the burden of proof as requiring nothing more specific than

reasons is controversial, as we shall see below. Thus, everybody accepting this basic, general idea means that it is necessary for accepting any other accurate, more specific interpretation of the burden of proof: the more precise interpretations imply or presuppose this broader, basic characterization.

98

On Presumptions, Burdens of Proof, and Explanations

provide some reason for her position. Two points must be made here. One is that, at first blush, this result appears consistent with standard accounts, since no standard account will deny that the Challenger must provide reasons upon rejecting a presumption. Another is that this interpretation of the deontic function preserves deontic similarity between cognitive and practical presumptions. In other words, regardless of whether the proposition is epistemically plausible (“Presumably(c), the Earth is globe-shaped”), or desirable to proceed on for non-epistemic reasons (“Presumably(p), John will be at the party”), the Challenger must provide reasons upon rejecting it. (t1) U: Presumably, p. (t2) C: Reject: “Presumably, p.” (t3) C: Reason: r. [discharging the BoR] If the BoR expresses the basic idea underlying the burden of proof, appears to be consistent with the standard characterization of deontic function, and preserves the similarity between the two types of presumption,106 why shouldn’t we define the burden of proof in terms of the BoR and conclude our analysis at this point? The answer is twofold. First, the less general conception of the burden of proof is more analytically useful since it raises essential theoretical questions and reveals differences between cognitive and practical presumptions. Second, philosophers and argumentation scholars (traditional theorists of presumptions included) typically adopt a less general conception. So, what exactly do they mean by this less general conception of the burden of proof?

3

4.2 Presumptions and the Burden of Arguing Let’s focus on the obligation to provide a reason, i.e., the burden of reasoning. This general dialectical obligation includes distinct sub-obligations, most notably the obligation to provide an argument, and the obligation to provide an explanation. Philosophers and argumentation scholars, however, typically contend that “[t]o have the burden of proof is to be rationally required to argue for or provide evidence for your position” (Dare & Kingsbury 2008, p. 503). In other words, the burden of proof is the obligation of an “arguer” (Johnson 2000, p. 194) to “argue” (van Eemeren & Houtlosser 2002, pp. 17) or present “arguments” (Walton 1988, p. 234; Freeman 2005, p. ix; Govier 2010, p. 175; van Laar & Krabbe 2013, p. 202). On this interpretation, the burden of proof is not just the burden of reasoning but, more accurately, the burden of arguing (BoA). The burden of arguing (BoA) represents the party’s dialogical obligation to provide a dialectically adequate, argumentative reason for a position (view).

106 

Contributing, thereby, to the overall coherency of standard accounts.

99

CHAPTER 3

To understand the nature of the BoA, we should keep in mind that the paradigmatic goal of arguing is to persuade the other party (Walton 1990; Blair 2012) or to resolve a difference of opinion (van Eemeren & Grootendorst 2004). To achieve this, an arguer should offer reasons that the other party, ideally, has already conceded (van Laar & Krabbe 2013, p. 212) or will most likely concede.107 I shall call these reasons argumentative reasons. In effect, the burden of arguing is the obligation to provide reasons that are acceptable to the other party, typically for the sake of persuading her. What does this mean for the deontic function of presumptions? Do presumptions impose the burden of arguing on the Challenger? According to standard accounts, they do. Traditional theorists contend that the Challenger must “disprove the proposition in question” (Walton 2014, p. 274), adduce “sufficient reason … against it” (Whatley 1963, p. 112), or “[show] that not-q” (Ullmann-Margalit 1983, p. 151). In dialogical contexts, however, the Challenger can hardly achieve these goals (only) by explaining her rejection. Rather, to disprove, defeat, or rebut a presumed proposition, the Challenger must present “an argument against it” (Pinto 2001, p. 4) or adduce “appropriately weighty counterarguments” (Rescher 2006, p. 16). Supposedly, cognitive and practical presumptions are the same in this respect: on standard accounts, the Challenger is obliged to provide an argument regardless of whether she rejects “Presumably(c), the Earth is globe-shaped” or “Presumably(p), John will be at the party.” (t1) U: Presumably, p. (t2) C: Reject: “Presumably, p.” (t3) C: Argumentative reason: a. [discharging the BoA.] To understand why standard accounts require qualifications, let’s start by analysing the deontic function of cognitive presumptions. Is the Challenger indeed obliged to present an argument at t3, immediately after rejecting the cognitive presumption? In an epistemic dialogue, it seems rather doubtful that the Challenger must immediately persuade and present a reason that is acceptable to the User. In the Flat Earth example, Steve might (also) be allowed to offer an explanation after rejecting Diane’s standpoint that “Presumably(c), the Earth is globe-shaped,” i.e., he might (also) be allowed to discharge the burden of reasoning by satisfying the burden of explanation (Rescorla 2009a; van Laar & Krabbe 2013). Why is this so? And what exactly is the burden of explanation?

4.3 Cognitive Presumptions and the Burden of Explanation To understand the nature of the burden of explanation, one thing must be clear: since Diane and Steve willingly engaged in a discussion about Earth’s shape, both should 107 

Scholars refer to this requirement by using different terms, such as “premise acceptability” (Johnson 2000, p. 194;

Govier 2010, p. 87), and “premise adequacy” (Goodwin 2001, p. 2).

100

On Presumptions, Burdens of Proof, and Explanations

have obligations to provide arguments. For instance, Diane carries the burden of arguing throughout the debate because she must make objections and cast doubt on Steve’s standpoint and arguments.108 However, since Diane can (successfully) discharge this burden only by presenting reasons acceptable to her Challenger, Steve’s rejection of “Presumably(c), the Earth is globe-shaped” puts her in a difficult position. Which reasons could she possibly use? If Steve rejects a proposition that, in normal circumstances, everyone in their right mind would concede, if well-known evidence cannot persuade him that the Earth is globeshaped, then what kind of reason could convince Steve of anything regarding this matter? Steve has shaken the very foundations of reasonable dialogue without providing any guidance on what grounds to continue. Without this kind of guidance, Diane will probably be unable to construct a persuasive argument. In the Flat Earth example, Diane is facing a complex problem. On the one hand, she cannot use many plausible, well-known, and widely-accepted reasons since they are dialectically inadequate (Steve implicitly dismisses them by rejecting “The Earth is globeshaped”). On the other hand, Diane cannot easily anticipate which reasons Steve will consider dialectically adequate: since Steve seems to refuse typical, well-known reasons, she can do nothing but guess what reasons might persuade him. The burden of explanation seeks to resolve the latter, transparency issue. Steve should “elucidate [his] position, thereby helping the original speaker isolate the relevant mutually acceptable premises” (Rescorla 2009a, p. 100),109 or, in van Laar & Krabbe’s (2013) words, Steve should explain the “motivation for [his] position” (p. 213) in order to provide “a strategic advice” (p. 212) or “strategic guidance” (p. 213). This strategic advice should give Diane a chance to make a persuasive, dialectically adequate argument.

3

The considerations related to the burden of explanation are different from the ones related to the burden of arguing. Specifically, providing an explanation “is not an attempt to convince the other and need not start from propositions conceded by the other” (van Laar & Krabbe 2013, p. 212). Thus, the burden of explanation does not require dialectically adequate, argumentative reasons, but merely explanatory reasons. 108  It is incorrect to say that (cognitive) presumption p exempts the User from bearing any burden of arguing throughout the entire discussion. In my view, as far as the User is concerned, a presumption does two things. First, immediately after the Challenger rejects p, it exempts the User from any burden of arguing. Second, in the discussion that follows, it exempts the User from the burden of arguing in favour of p (despite p being the User’s standpoint), but it does not exempt her from presenting arguments and objections against the Challenger’s views. In other words, presumptions could, perhaps, work in the proponent’s favour by allowing her to win the mixed discussion by acting like a successful critic. 109  Although Rescorla (2009a) seems to associate explanations with ‘rebuilding’ common ground (rapprochement), I believe that the purpose of explanation is more general. That is, even if explanations are legitimate primarily when common ground is at stake, the purpose of explaining might be to provide any premises that the proponent might use to make a persuasive argument. Sometimes, these premises can be acceptable only to the Challenger and, thus, cannot represent new common ground. However, the proponent can still make use of them to make a persuasive case. Also, note that I deal with only one particular function of explanation, namely the function of improving “directiveness” (van Laar & Krabbe 2013) and thereby of optimizing conditions for the constructive development of dialogue. However, one could use explanations for many other purposes. For instance, one could also offer inadequate, explanatory reasons to end any further discussion about a particular matter.

101

CHAPTER 3

The burden of explanation (BoE) represents the party’s dialogical obligation to provide an explanatory reason for a position (view).110 For instance, after rejecting “Presumably(c), the Earth is globe-shaped,” Steve can offer the following explanation: “Our government fabricated the evidence that the Earth is globeshaped.” This explanatory reason is neither persuasive to nor conceded by Diane, but it is helpful in resolving the aforementioned transparency issue: it reveals to Diane what she needs to attack, as well as how to attack Steve’s position adequately.111 (t1) U: Presumably(c), p. (t2) C: Reject: “Presumably(c), p.” (t3) C: Explanatory reason: e. [discharging the BoE] Notice, for instance, that the previous explanation makes Steve committed to a more general claim, namely that “Our government fabricates evidence about the shape of the Earth.” After realizing this, Diane might exploit Steve’s commitment as a premise of her adequate counterargument. Imagine that she decides to construct an alternative conspiracy theory: “But our government could have fabricated the Flat Earth theory just to keep the public away from the important stuff. The Flat Earth movement had the best media coverage on the day when our government passed the controversial law on public health. Don’t you find this suspicious?” To be sure, Diane’s counterargument does not show that the Earth is globe-shaped, but it might show that both “The Earth is globe-shaped” and “The Earth is disc-shaped” must be equally acceptable to Steve given his commitment that governments fabricate evidence. Since Diane has a presumption in her favour, this is all she needs to do to (successfully) discharge her burden of arguing and to prevent Steve from winning the discussion. In the Flat Earth example, Steve’s explanation, however bizarre, plays a constructive dialectical role because it helps Diane to participate in the discussion fruitfully. Also, it nicely coheres with the widely-accepted, Gricean view that argumentative discussion 110  As Godden notes, explanations are always permitted: “there is no prohibition, in standard ‘symmetrical’ dialectical games, against arguers motivating or explaining their doubts of other’s standpoints” (2020, p. 4). But what makes explanations obligatory once presumptions are rejected? According to van Laar & Krabbe (2013), the opponent owes the proponent an explanation because, by rejecting the initial presumption, she changes the original dialogical setting and makes it very difficult for the proponent to make a persuasive case. Similarly, Rescorla (2009a, pp. 97-98) argues that the opponent must try to achieve rapprochement (i.e., try to isolate mutually acceptable propositions) because this is constitutive of a reasoned discourse. Explanations are a means to achieve this goal, and they help the proponent to discharge her burden of proof successfully. In my opinion, the Challenger must present the explanation only if the User requested the explanation in the previous turn. Otherwise, after rejecting a cognitive presumption, the Challenger is obliged to give either an argument or an explanation. 111 

After Steve’s explanation, Diane knows that to cast doubt on “The Earth is disc-shaped,” she must at some

point make a persuasive refutation of “Our government fabricated the evidence that the Earth is globe-shaped.” More importantly, she can use Steve’s explanation to derive adequate reasons for her arguments.

102

On Presumptions, Burdens of Proof, and Explanations

is a cooperative enterprise. For these two reasons, explaining seems like a reasonable dialectical choice at t3. But if explaining is dialectically reasonable, then how can Steve be obliged to provide an argument, as standard accounts of deontic function seem to suggest? If he is allowed to discharge the burden of explanation, how can he carry the burden of arguing (proof)? Ultimately, do cognitive presumptions asymmetrically allocate the burden of arguing (BoA) or the burden of explanation (BoE) at t3? I believe that cognitive presumptions allocate neither of these burdens. Technically, when the Challenger rejects a cognitive presumption, she is not immediately obliged to provide an argument (in order to defend her position), and she is not immediately obliged to give an explanation (in order to provide strategic guidance). Since providing an argument and providing an explanation are both reasonable moves, I believe that the Challenger’s obligation is rather a disjunction: when she rejects a cognitive presumption at t2, she is obliged either to provide an argument or to explain the rejection at t3. In other words, the Challenger incurs the burden of reasoning (BoR), and she can discharge this burden in a more competitive (by discharging the BoA) or a more cooperative fashion (by discharging the BoE).

3

Figure 1: A profile of dialogue: cognitive presumption

To be sure, the Challenger might become obliged to argue at t4 if the User requests an argument at t3, but this would mean that the Challenger’s burden is conditional upon the User’s speech act.112 As long as the BoA is unconditionally allocated at t3, I believe that the standard accounts are incorrect: at t3, the Challenger does not carry the BoA, but the more general BoR. This does not mean, however, that the Challenger is exempted from the global burden of proof: if her final aim is to persuade the proponent and win the discussion, then, ultimately, she will be obliged to defend her standpoint with (sufficient) argument. But at the local level, at t3, immediately after rejecting the presumption, the Challenger is permitted to present an explanation, as well as an argument (and obliged to present either explanation or argument). 112  As briefly mentioned at the beginning of Sect. 4, it is unclear whether standard accounts of deontic function accept this idea.

103

CHAPTER 3

In conclusion, cognitive presumptions asymmetrically allocate the burden of proof at t3 only if the burden of proof is interpreted in terms of the BoR. As soon as we accept the usual, more specific interpretation of the burden of proof in terms of the BoA, standard accounts of deontic function require revisions: cognitive presumptions do not asymmetrically allocate the burden of proof (at t3). But what does this mean for the relationship between cognitive and practical presumptions? Are challengers obliged to offer either explanations or arguments after rejecting practical presumptions, or do practical presumptions, unlike cognitive ones, asymmetrically allocate the burden of proof in the strict sense of the BoA?

4.4 Practical Presumptions and the Burden of Explanation I argue that practical and cognitive presumptions have distinct deontic functions. The differences, however, are subtle. To make this clear, I need to distinguish between two views that differently specify when one can impose the burden of explanation on the challenger of a practical presumption. I call them the stricter view and the looser view. According to the stricter view, the burden of explanation should only suffice when the opponent challenges a common ground proposition. Since the rejection of a widely-accepted (or mutually-accepted) proposition “counts as a substantial change of the nature of the dialogue” (van Laar & Krabbe 2013, p. 210), the Challenger must make sure that the new conditions for making a persuasive argument are transparent to the proponent. Nevertheless, none of these considerations directly apply to practical presumptions since, on the standard view, practical presumptions are not common ground propositions. Instead, they are “new intellectual resources” used at some late stage of deliberation to “proceed with our undertakings” (Godden 2017, p. 487). Accordingly, the Challenger does not change the nature of the dialogue by rejecting a practical presumption, and the User’s task of recognizing dialectically adequate reasons should not be especially demanding. Let’s illustrate this on the Party example. Alice: I would like to go to the party tonight. Mark: I would like to go too. But if it is uncertain whether John will be there, I would rather stay home. His presence might entirely ruin our evening. Alice: I agree. As long as there is a reasonable chance that John will come to the party, we should err on the side of caution and skip it. It is not worth the risk. Mark: Can we somehow check whether he will attend? Alice: I asked some colleagues, but they never texted me back. Anyway, we should decide quickly. The train is about to leave. Mark: Isn’t John out of town right now? He was bragging about his upcoming trip to Spain, remember? 104

On Presumptions, Burdens of Proof, and Explanations

Alice: Yes, but he must be at work the day after tomorrow, so… (t1)

Mark: If he is back in town, he will probably be at that party. Since we are uncertain about his whereabouts, let’s just skip the party this time. [= “Presumably(p), John will be at the party.”]

(t2)

Alice: Well, I think we should go.

By rejecting “Presumably(p), John will be at the party” at t2, Alice makes a surprising move. Although, technically, she does not challenge the common ground proposition,113 she rejects the position that appears to follow from her commitments.114 Alice should then provide a reason for her rejection, but is she allowed to provide (only) an explanatory reason? Does it suffice to discharge the burden of explanation? If only attacks on common ground propositions might incur the burden of explanation, then it does not suffice to impose on the Challenger the mere burden of explaining her position. Namely, the purpose of explaining is to offer strategic guidance, but when a practical presumption gets challenged, the User does not seem to need any strategic advice. Even before Alice gives any reason at t3, Mark has a fairly clear idea of what might persuade her to skip the party. Starting from the belief that Alice still wants to avoid John, Mark can use any reason proving that John will definitely (or most probably) come to the party. He can perhaps call John and after receiving confirmation from him, tell Alice the bad news, or provide evidence that John is already in town. Of course, Mark might be unable to provide an adequate, persuasive argument at some particular point, but this is an entirely different matter. What is crucial is that he has a good idea of what, in principle, might constitute an adequate, persuasive argument and that, consequently, he does not need strategic advice on how to resolve the transparency issue. Since this need underlies the burden of explanation, explaining becomes irrelevant and, therefore, does not suffice in the Party example.

113 

3

There are at least three reasons why “Presumably(p), John will be at the party” does not constitute the common

ground. First, “John will be at the party” is certainly not a widely-accepted proposition supported by many epistemic sources (like “The Earth is globe-shaped”). Second, “John will be at the party” is not a shared concession, i.e., the proposition accepted by Alice and Mark at the beginning of the dialogue. Third, even when we interpret this proposition in the pragmatic, action-orienting sense of “We should act as if John will be at the party,” Alice and Mark do not accept it at the beginning of the dialogue. At best, they accept a (presumptive) rule “If it is uncertain whether John will be at the party, then we should act as if John will be at the party” and then, at t1, Mark derives the presumption based on this rule and its antecedent (basic fact) “It is uncertain whether John will be at the party.” In short, a practical presumption is not a common ground premise that should be accepted before the main discussion, in order to enable a desirable start of the discussion. Rather, it is a tentative conclusion drawn at some later point in the discussion, to enable a desirable end of deliberation (i.e., avoiding greater harm in the context of evidential uncertainty and pressure to make a timely decision). 114 

That is, for all that Mark knows, Alice seems committed to both the presumptive rule and the basic fact at t2.

Consequently, one would expect that she will not reject the presumption at t2.

105

CHAPTER 3

Thus, on the stricter view, explanation represents a means without an end: it is designed to resolve a dialectical problem that, once a practical presumption gets rejected, does not arise. For this reason, it is insufficient for the Challenger to explain her rejection; she must also present an argument. (t1)

U: Presumably(p), p.

(t2)

C: Reject: “Presumably(p), p.”

(t3)

C: Argumentative reason: a. [discharging the BoA]

Figure 2: A profile of dialogue: practical presumption (stricter view)

According to the looser view, the burden of explanation also suffices when the challenged proposition is not part of the common ground. To be sure, proponents will especially need guidance when the common ground propositions get rejected, but dialectical rules should also permit explanations when common ground is not at stake, and the discussion is already under way. In principle, offering strategic advice is a cooperative move under any circumstances, and the dialectical rules should not penalize the Challenger for being too cooperative. They must allow explanations and strategic advice even when the Challenger’s explanations and guidance are, perhaps, unnecessary. But are practical presumptions, then, different from cognitive presumptions? If both types permit explanations as well as arguments, then, immediately after they are rejected, all presumptions impose the general burden of reasoning on their opponents. Do cognitive and practical presumptions, under the looser interpretation of the burden of explanation, have the same deontic function? I propose the following answer: within the framework of an abstract dialectical model, insensitive to the unique circumstances in which practical presumptions operate, they do share the same deontic function. However, within the framework of a dialectical model that takes the special circumstances of uncertainty and deliberative pressure into account, the deontic functions should differ. At t3, immediately following the Challenger’s rejection of the practical presumption, explanations are permissible, but not dialectically desirable. That is, although the Challenger’s explanations do not seem to violate any ground rules for reasonable discussion, they hardly contribute to the optimal resolution of deliberation in circumstances of uncertainty and pressure. Hence, the rules of dialogue should strongly encourage the Challenger to provide an argument at t3. Since this does not apply to cognitive cases, we can do justice to the differences between cognitive and practical presumptions even on the looser view. In the profile below, I illustrate the Challenger’s permitted, yet suboptimal response with a curved line.

106

On Presumptions, Burdens of Proof, and Explanations

Figure 3: A profile of dialogue: practical presumption (looser view)

But what makes explanations suboptimal in practical cases, like the Party example? Why exactly should rules encourage Alice to support “We should go to the party” with argument rather than explanation? I already mentioned one reason: offering strategic guidance to Mark is simply not needed, since Alice does not reject any common ground propositions. The more interesting reasons, however, are related to the lack of impact, or even the negative impact that the Challenger’s explanations could have on the resolution process. Usually, the Challenger’s explanations might be irrelevant or even detrimental to an optimal resolution of deliberation. How is this possible?

3

Recall that explanatory reasons are not aimed at persuasion and that after rejecting a presumption at t2, Alice might say the following: “The tea leaves tell me that John will not come to the party.” Since Mark does not believe in the prophetic properties of tea leaves, her explanation is not persuasive, although it seems dialectically permissible.115 To see why her explanation is not dialectically desirable, we must appreciate two unique deliberative limitations underlying the Party example: (1) uncertainty and (2) time pressure. Since Mark and Alice wish to avoid greater harm (seeing John), they will attend the party only if they are certain enough that John will skip it, and at t2, they seem uncertain about John’s whereabouts. Also, the time left for gathering evidence about John’s whereabouts is limited, and Mark and Alice must decide quickly (since the train is about to leave). I believe that Alice’s explanation is irrelevant, since it cannot affect their uncertainty and, ipso facto, it cannot change their default course of action.116 When evidence is uncertain, practical presumptions provide a default course of action, which, in the Party example, is skipping the party: if it remains (dialectically) uncertain whether John will 115  In my view, from a dialectical standpoint, Alice does not commit any argumentative fallacy. Even more importantly, she is cooperative and transparent. She cooperates with Mark by presenting evidence that she considers relevant and, presumably, the only additional evidence she is capable of presenting at t3 (thereby offering strategic advice). 116 

In the context of dialogue, certainty is understood in a dialectical sense: proposition p is certain if, among other

things, both parties agree that it is (likely to be) true. Otherwise, it is treated as uncertain. Notice that this does not exclude objective epistemic considerations. I do not claim that mutual agreement that p is (likely) true is a sufficient condition to render p dialectically certain; instead, I claim that it is a necessary condition to render p dialectically certain.

107

CHAPTER 3

come to the party at the point in time when Mark and Alice must go to the train station, then Mark and Alice will skip the party. To take some other action, Mark and Alice must agree that it became sufficiently certain that John will not be at the party. But since reading tea leaves will not persuade Mark, the proposition “John will skip the party” will remain dialectically uncertain, and the original presumption will remain in place. As a result, the proposed explanation will not affect the outcome of deliberation: Mark and Alice will skip the party, i.e., they will do what they would have done even if the explanation had not been offered. So why offer explanations, even if they are allowed? In the described circumstances, explaining seems pointless.117 Not only is providing an explanation irrelevant, but it is also detrimental. Practical presumptions arise when time is limited, representing an especially valuable resource. In these special circumstances, it is usually not reasonable to spend time (only) on giving explanations since they, as we have seen, can hardly change the default outcome of deliberation. Since parties want to improve their chances of arriving at the best decision, which might be different from the default decision, they should probably spend their limited time in a more constructive way. The first and obvious choice is to present and discuss arguments that are already at hand. If, at t3, Alice has both an explanation and an adequate argument up her sleeve, then she should present the argument. By doing so, Alice will have a better chance to persuade Mark, affect dialectical uncertainty, and move deliberation towards optimal resolution. The second and less obvious choice is to spend time on finding arguments that are not already at hand. Suppose that, at t3, Alice has a choice to make: provide an explanation, or search for reasons that might be relevant for making the best decision. For instance, instead of spending time discussing tea leaves, Alice might simply call John and ask him about his whereabouts. If Mark and Alice consider John’s testimony reliable, acquiring this information will have a direct impact on reaching an optimal decision in the Party example. Does this render the looser view implausible? If (only) explaining is typically irrelevant in circumstances of uncertainty and detrimental in circumstances of pressure, should (only) explaining even be allowed? In my opinion, this conclusion is too extreme. Explanations should not be banned by dialectical rules because, at least sometimes, they can be persuasive/adequate to a limited degree and represent weak arguments. That the User does not concede the Challenger’s explanatory reason could simply mean that she finds this reason adequate to a minimal degree. But in situations of evidential uncertainty, where any piece of evidence (however weak or inadequate) might make a difference for arriving at the best decision, explanations with limited persuasive force might be 117 

What happens if Mark uses information about tea leaves to persuade Alice that John’s whereabouts are still

uncertain? Does explanation, perhaps, affect the outcome of deliberation by giving strategic guidance to Mark? I believe that it does not. The presumption remains in place as long as Mark maintains his scepticism about the reliability of tea leaves and John’s whereabouts, and this does not seem to depend on whether his future attempt to persuade Alice, by exploiting her explanation concerning tea leaves, is successful or not.

108

On Presumptions, Burdens of Proof, and Explanations

both relevant and constructive. Hence, as far as rejections of practical presumptions are concerned, explanations should be permitted but not recommended (compared to adequate arguments) since, at least sometimes, they might contribute to something more than providing strategic guidance. To sum up, explaining may come with a high opportunity cost and be detrimental for arriving at the best decision. When the Challenger rejects a practical presumption, explaining is usually a waste of time, in both figurative and literal terms. Crucially, notice that similar concerns do not easily apply to cognitive presumptions. However bizarre, Steve’s explanation “Our government fabricated the evidence that the Earth is globe-shaped” affects the further development of the dialogue in the Flat Earth example, and since it provides much-needed guidance to Diane, its influence is quite positive. In addition, it seems that Diane and Steve do not need to worry about approaching deadlines. Technically, their debate about the shape of the Earth can last for years, and Steve is not irresponsibly wasting time by offering a single explanation (at least not in the sense that is specific to the Party example). In effect, the two types of presumptions have different deontic functions.

3

The deontic function of cognitive presumptions: After they are rejected at t2, cognitive presumptions allocate the burden of reasoning (BoR) to the Challenger at t3. This means that the Challenger is obliged to discharge either the burden of arguing (BoA) or the burden of explanation (BoE). The deontic function of practical presumptions: After they are rejected at t2, practical presumptions, in principle, allocate the burden of reasoning (BoR) to the Challenger at t3. However, typically, given the circumstances of uncertainty and deliberative pressure, the Challenger should discharge the BoA: dialectical rules should permit explanations and arguments at t3, but strongly recommend arguments over explanations. These characterizations of deontic functions are not complete. First of all, they only capture the Challenger’s rights and obligations. But what about the proponent? Is the User of a presumption indeed exempted from the burden of proof? In which sense of the term, and under what conditions? Can the Challenger’s explanation shift the burden of proof on the User, or only an argument would suffice? Does the argument need to be sufficiently strong, or only adequate? Moreover, the present characterizations do not explore the conditionality of the Challenger’s obligation. Is the Challenger immediately obliged to give reasons at t3, or does her obligation at t4 depend on the User’s request for reasons at t3? And who gets to choose whether argumentative or explanatory reasons should be presented: the User or the Challenger? The present chapter starts from the assumption that standard accounts propose an unconditional allocation of the burden of proof, but this must be explored in more detail. Finally, the crucial concept of probative asymmetry, or ‘asymmetrical allocation’ has not been explored at all. 109

CHAPTER 3

However, even at this preliminary stage, our results show that standard accounts of presumption require qualifications. On the one hand, if standard accounts insist that presumptions allocate the burden of proof asymmetrically, then they seem committed to a rather loose and ambiguous notion of the burden of proof: whereas cognitive presumptions reverse the burden of proof in the sense of the burden of reasoning, practical presumptions reverse the burden of proof in the sense of the burden of arguing (at least in optimal scenarios). On the other hand, if standard accounts accept the natural conception of the burden of proof, as I believe they do, then their characterization of deontic function is inaccurate: immediately after they are rejected, only practical presumptions (should) asymmetrically allocate the burden of proof. This reveals that cognitive and practical presumptions potentially create different patterns of dialectical interaction.

5 Conclusion According to standard accounts, presumptions are dialectically privileged, yet defeasible propositions: if the Challenger rejects a presumption, she is supposed to carry the burden of proof, and the presumption is acceptable until the burden of proof is (successfully) discharged. Standard accounts acknowledge that there are various types of presumptions, but they treat deontic function as a shared dialectical feature or a kind of common denominator. In short, once rejected, all presumptions are supposed to impose the burden of proof on the Challenger. In this chapter, I analysed the deontic function by taking into account (a) different types of presumption, and (b) distinct conceptions of the burden of proof. First, I argued that presumption, taken in the abstract, does not impose the burden of proof on the Challenger immediately after it is rejected. Rather, the Challenger incurs the general burden of reasoning. Since the burden of proof is typically a narrower conception than the burden of reasoning, this conclusion is not entirely in line with standard accounts. Second, I argued that cognitive and practical presumptions distribute different dialectical obligations. To be sure, they both impose some kind of burden of reasoning on the Challenger, but whereas cognitive presumptions require either arguments or explanations, practical presumptions seem to require (or at least strongly recommend) arguments. Thus, at the level of concrete dialogical implementation, deontic uniqueness, proposed by standard accounts, does not hold. Different types of presumptions have distinct deontic functions. The results of this chapter are relevant for argumentation theory, in particular, and philosophy, in general. In argumentation theory, they contribute to the normative study of the Challenger’s critical reactions, i.e., the “burden of criticism” (see Krabbe & van Laar 2011; van Laar & Krabbe 2013). In philosophy, they contribute to the ongoing discussion between dialectical foundationalists and dialectical egalitarianists. For obvious reasons, they contribute to the dialectical study of presumptions. Nevertheless, the present results are provisional, and much additional work is needed to test their tenability.

110

On Presumptions, Burdens of Proof, and Explanations

Acknowledgments: The earlier versions of this chapter were presented at the Annual PhD Conference Groningen (Groningen, September 25 2019), as well as at the 12th OSSA conference Evidence, Persuasion & Diversity (Windsor, June 3-6 2020). I thank the organizers of these conferences, as well as the participants for their constructive criticisms and recommendations. I am especially indebted to my supervisors dr. Jan Albert van Laar and dr. Barteld Kooi, my OSSA commentator dr. David Godden, the reviewers and editors of Informal logic (especially dr. J. Anthony Blair), and Karolina Kudlek. Finally, I thank the chair of the committee, dr. Sally Jackson, and the members of the committee for awarding the earlier version of this chapter with the ‘J. Anthony Blair Prize’ for an outstanding student paper.

3

111

4

CHAPTER 4 On Presumptions, and Asymmetrical Allocations of Burden of Proof

CHAPTER 4

Symmetry is what we see at a glance; based on the fact that there is no reason for any difference. —Pascal, Pensées

Why cover the same ground again? ... It goes against my grain to repeat a tale told once, and told so clearly. —Homer, The Odyssey

114

On Presumptions, and Asymmetrical Allocations of Burden of Proof

On Presumptions, and Asymmetrical Allocations of Burden of Proof Abstract: Traditionally, presumption p is associated with an asymmetrical allocation of the burden of proof. Supposedly, while p’s Challenger must defend her view, p’s User is dialectically privileged. But what is the meaning of ‘asymmetrical allocation’? Is the User entirely exempted from a probative burden, or is she still committed to present an argument (however weak)? Must the Challenger provide a sufficient argument, or can she give a minor one? And what types of normative considerations underlie these questions? This chapter, first, offers a comprehensive and systematic account of distinct kinds of ‘asymmetrical allocation.’ Second, it investigates the relations between asymmetrical allocations and presumptions. It shows that practical and cognitive presumptions entail distinct asymmetries, or entail identical asymmetries but in different dialogical settings. Consequently, the presumption’s unique deontic function—as portrayed in standard accounts—is fragmented. ‘Asymmetrical allocation’ cannot truly serve as a common denominator that conceptually unifies all presumptions (at least not without qualifications).

4

Keywords: asymmetrical allocation, burden of proof, cognitive presumption, commitment, practical presumption, persuasion burden, production burden, transparency.

1 Introduction One central question of argumentation studies is how to regulate a reasonable dialogue. Argumentation scholars try to identify rules of argument production, analysis, and evaluation that systematically further the goal of a critical discussion (van Eemeren & Grootendorst 2004) or some other dialogue type (see Walton & Krabbe 1995; Walton 1998). Thus, the goal of dialogue generates the rules, and thereby parties’ permissions, obligations, and responsibilities (specified in the rules). According to the “Principle of Burden of Proof” (Krabbe 1995, p. 255), the “obligation-to-defend rule” (van Eemeren & Grootendorst 2004, p. 191), or the so-called “defence norm” (Rescorla 2009a, p. 88), a party is obliged to “furnish support for the thesis” (Johnson 2000, p. 194) or “defend the standpoint once challenged to do so” (van Eemeren & Houtlosser 2002, p. 17) in order to “show that one of her claims is correct” (Bermejo-Luque 2011, p. 193). But does the Principle of Burden of Proof (always) apply to both discussants when they advance contradictory standpoints—i.e., when a difference of opinion is mixed (van Eemeren & Grootendorst 2004, p. 60)—or are there unique standpoints that, though challenged, do not require defending? In this chapter, I analyse dialectically privileged propositions that, immediately after they are challenged, supposedly, do not require (sufficient) defence. The analysis focuses on one special type of dialectically privileged proposition; namely, a presumption. 115

CHAPTER 4

On standard accounts, the proponent is not required to defend a presumption p unless the challenger sufficiently justifies p’s rejection (Whately 1963; Rescher 1977, 2006; Freeman 2005; Walton 2014).118 Although contemporary scholarship offers conflicting typologies, it usually distinguishes practical from cognitive presumptions (see UllmannMargalit 1983; Rescher 2006; Godden & Walton 2007; Lewiński 2017; Godden 2019). A practical presumption is portrayed as an epistemically uncertain proposition that an agent should proceed on so as to avoid greater harm in deliberation (UllmannMargalit 1983; Godden 2017). In criminal law, if it is uncertain whether the accused is guilty or innocent, the court must presume (proceed as if) she is innocent since sentencing an innocent person is considered more harmful than letting a guilty person free. In nonlegal deliberation, if Larry is uncertain whether the weapon is loaded, he should presume (proceed as if) it is loaded because this avoids potentially fatal consequences (Walton 1988). So, from a dialectical viewpoint, the proposition the acceptance whereof avoids greater harm should not entail the burden of proof: its proponent should not be obliged to present (sufficient) reasons unless the challenger sufficiently supports its rejection. A cognitive presumption is portrayed as a plausible proposition that an agent should accept to facilitate epistemic inquiry. Rescher defines it as “the most plausible of rival alternatives” (2006, p. 39), and Freeman contends that we identify it by considering the epistemic source (2005, pp. 41-42). For instance, Diane should recognise that the scientific consensus (a reliable interpersonal source) makes “The Earth is globe-shaped” the most plausible truth-candidate. In effect, “The Earth is globe-shaped” represents a cognitive presumption, and the burden should rest with the party who rejects it (e.g., flat-Earther Steve). Accordingly, from a dialectical viewpoint, the most plausible proposition should not entail the burden of proof: its proponent should not be obliged to present (sufficient) reasons unless the challenger sufficiently supports its rejection. Practical and cognitive presumptions differ in several respects (see Bodlović 2021), but, supposedly, they share the same deontic function. They are said to “shift” (Pinto 2001; Freeman 2005; Walton 2014), “reverse” (Rescher 2006; Walton 2019; Godden 2019), “differentially allocate” (Dare & Kingsbury 2008), and “unevenly distribute” (Aijaz et al. 2013) the burden of proof. Although scholars have used various expressions to denote this unique function, the underlying idea—that presumptions benefit proponents and burden challengers—is essential on all standard accounts. On the one hand, it holds all presumptions together and, on the other, separates them from (regular) claims, assumptions, stipulations, presuppositions, and hypotheses (see Walton 1993; Freeman 2005; Rescher 2006; Godden 2017). In a nutshell, the unique deontic function is a cornerstone of presumption’s conceptual coherency. 118  The other class of privileged propositions is called ‘fixed concessions.’ Those propositions are “unretractable throughout the discussion at hand” (van Laar & Krabbe 2013, p. 210), so, unlike presumptions, the opponent is not allowed to challenge them under any circumstances, and the proponent cannot become obliged to defend them. For instance, in the normal, everyday context the reasonable opponent should not be allowed to challenge that triangle has three angles, and the proponent of “The triangle has three angles” should never be obliged to offer arguments for this claim.

116

On Presumptions, and Asymmetrical Allocations of Burden of Proof

The unique deontic function of presumption (standard view): A presumption (1) allocates (2) the burden of proof in an (3) unconditional and (4) asymmetrical manner.119 But do different presumptions asymmetrically allocate the burden in a comparable sense, and in the identical dialogical settings? This chapter tackles the following question: Assuming that presumptions unconditionally and asymmetrically allocate the same kind of burden,120 what meaning should be assigned to ‘asymmetrically’, both in the case of practical and cognitive presumptions? Is asymmetry linked to doxastic or dialectical requirements? If the asymmetry is doxastic, is it associated with unequal standards of proof, meaning that the presumption’s proponent must prove that her proposition is worth considering, while the challenger must prove that her proposition is justified belief? If the asymmetry is dialectical, is it a matter of presentation, meaning that the same evaluation standard applies to both parties, but only the presumption’s challenger must present sufficient reason? Or is it linked to the strictness of commitment, meaning that the challenger is obliged to provide sufficient reasons while the proponent is only recommended to do so? To develop a more sophisticated theory of presumption, we must address these questions. Scholars recognise different types of asymmetrical allocation (see, e.g., Hahn & Oaksford 2007; Dare & Kingsbury 2008; Aijaz et al. 2013), but usually discuss them in passing, without offering a systematic analysis. This chapter’s conceptual objective is to develop a comprehensive account of ‘asymmetrical’ by organising the familiar kinds of asymmetry into a systematic network, and exploring neglected kinds of asymmetrical allocation (e.g., the meaning of ‘commitment strictness asymmetry’). The chapter’s typological objective is exploring the relationship between practical and cognitive presumptions by using the results of previous conceptual analysis. The third, dialectical objective is to enhance understanding of how presumptions should operate in a discussion. This seems necessary for assessing and improving the standard accounts of presumption.

4

The chapter concludes that practical and cognitive presumptions entail different asymmetries, or entail the same asymmetries but in distinct dialogical settings. Hence, the unique deontic function—a presumption’s distinctive feature—is fragmented and can hardly serve as a common conceptual denominator of all presumptions. Moreover, although standard accounts define presumptions in terms of burden asymmetries, such 119 

Bodlović calls this the “standard deontic function of presumption” (2020b, p. 272). Here, I replace ‘standard’ with

‘unique.’ The reason is twofold. First, saying that something is the standard deontic function might entail that presumptions have some other, non-standard deontic functions. Since mainstream accounts do not characterise presumptions in this way, such a label is not optimal. Second, the label ‘deontic function’ fails to provide information, stressed in many accounts, that presumptions distribute obligations in an unusual way. It fails to distinguish presumptions from regular assertions, assumptions, or hypotheses because they also distribute obligations. For these reasons, I use ‘the unique deontic function of presumption.’ 120  This is, literally, an assumption. As Bodlović (2020b) has recently argued, two types of presumption asymmetrically allocate distinct types of the burden of proof: ideally, practical presumptions allocate the burden of arguing (the obligation to produce an adequate argument), while cognitive presumptions allocate the burden of reasoning (the obligation to produce either an adequate argument or an explanation) on their challengers.

117

CHAPTER 4

asymmetries seem partly determined by the presumption’s contingent features. For instance, the presumption’s proponent is exempted from arguing (unlike the challenger) due to the presumption’s plausibility or pragmatic desirability, but also the accessibility or transparency of the presumption’s justification. However, strictly speaking, the justification’s accessibility or transparency is not a presumption’s essential feature. In Section 2, I define the ‘burden of proof’ and ‘allocation.’ In Section 3, I explore the relationship between presumptions and ‘binary asymmetries,’ i.e., allocations where one proponent is entirely exempted from supporting her position. In Section 4, I analyse ‘degree asymmetries,’ i.e., allocations where both proponents must support their positions, but one proponent has a more demanding burden. The conclusion offers a summary of the most important results.

2 Preliminary Clarifications Most generally, presumptions allocate the burden asymmetrically in the sense that the presumption’s challenger has a heavier burden of proof than the presumption’s proponent. But before analysing the parameters affecting the weight of the burden and generating asymmetries, preliminary clarifications are needed. What exactly is meant by the ‘burden of proof’ and ‘allocation?’121

2.1 Kinds of Burden of Proof There are three kinds of distinctions concerning the burden of proof. First, argumentation theorists distinguish a “burden to have good reasons” from a “burden to present some reasons” (Räikkä 1997, p. 467). Second, there is a burden to present any reason, as opposed to a burden to demonstrate a dialectically adequate reason, i.e., an argument. Third, a burden to present a relevant argument of any strength differs from a burden to give sufficient argument. While the first two distinctions are needed to specify the chapter’s scope, the third distinction is needed to answer its central question. The obligation to have reasons arises when agents have “some sort of positive attitude, or stance, towards the proposition” (Aijaz et al. 2013, p. 264). This attitudinal burden follows from a basic norm of rationality that “people should think and do things for adequate reasons” (p. 262). By contrast, the obligation to present or provide reasons follows from “dynamics of communication or deliberation” (p. 280). This dialectical burden (p. 262) arises because agents “should reason and argue effectively and efficiently” (p. 262) in the dialogical context. When the opponent challenges the proponent’s 121 

To develop a fully-fledged account of deontic function, we also need to understand the meaning of unconditional

allocation. In the present analysis, the notion of (un)conditionality does not play a significant role. However, the burden of proof is unconditionally allocated if the party that has the burden must produce (sufficient) reasons regardless of whether the other party explicitly demands them.

118

On Presumptions, and Asymmetrical Allocations of Burden of Proof

standpoint in a critical discussion, providing reasons represents a necessary and effective communicative means to persuade the challenger. Consequently, the proponent must comply with the division of dialectical burden (van Eemeren & Houtlosser 2002, p. 17). Although the attitudinal burden is helpful in explaining some contrasts between different types of presumptions, the present chapter centres on relationships between presumptions and dialectical burden allocations. However, the dialectical burden is a wide concept. Broadly speaking, it denotes a burden of reasoning, i.e., a party’s commitment to provide any reason, either argumentative or explanatory. In a more specific sense, it denotes a burden of arguing, or an arguer’s obligation to provide an argument. These obligations differ due to a requirement of dialectical adequacy (see Bodlović 2020b, pp. 272-275). Since arguing typically aims at persuasion, an arguer must give reasons acceptable to her opponent (van Eemeren & Houtlosser 2002, p. 20; van Laar & Krabbe 2013, p. 212). Yet, when reasoning does not aim at persuasion, argumentative adequacy is unnecessary; for instance, if the party only wishes to explain her view to the interlocutor, she can provide non-persuasive reasons (2013, p. 212). Despite the analytic importance, this chapter sets the distinction between the burdens of reasoning and arguing aside, focuses on the dialectical burden of arguing, and explores asymmetrical allocations regarding this type of burden.122

4

We can further specify our research question: If a presumption unconditionally and asymmetrically allocates the dialectical burden of arguing, what meaning should we assign to ‘asymmetrically,’ both in the case of practical and cognitive presumption? Before tackling this question, we must know that the burden of proof is “notoriously complex and ambiguous” (Prakken & Sartor 2009, p. 223) even in this narrow sense. In legal scholarship, J. B. Thayer (2019) identified two distinct obligations corresponding to the burden of arguing: the burden of production, and the burden of persuasion.123 On the standard view, the burden of production is a dialogical party’s obligation to provide prima facie argument for the challenged claim, for the sake of “avoiding claims unsupported by evidence” (Hahn & Oaksford 2007, p. 43). If the party does not support a challenged claim, she loses the discussion on that statement. But although not-fulfilling the production burden entails losing, fulfilling it does not entail winning: producing a prima facie argument is necessary to move the dialogue forward but is not sufficient, at least not always, to 122  For the analysis of the deontic function of presumptions, taking into account the difference between the burdens of reasoning and arguing, see Bodlović (2020b). 123 

Notions that are equivalent or very similar to the burden of persuasion are “risk of non-persuasion,” “probative

burden,” “probative burden of an original assertion,” “initiating burden,” “global burden,” “macro burden,” “objective burden,” and “the duty/burden of establishing.” Notions that are equivalent or very similar to the burden of production are “evidential burden,” “burden of producing/adducing/bringing forward evidence,” “burden of further reply (in the face of contrary considerations),” “proof of going forward,” “local burden,” “micro burden,” and “subjective burden” (see Cleary 1959; Rescher 1977, 2006; Prakken, Reed, and Walton 2005; Hahn & Oaksford 2007; Prakken & Sartor 2009; Thayer 2019; Ilbert 2019).

119

CHAPTER 4

win the discussion on a challenged claim.124 Moreover, the production burden is allocated and verified at different points in the discussion: the proponent must produce evidence whenever her statement becomes challenged (Prakken & Sartor 2009, p. 228; p. 247). The burden of persuasion is a party’s obligation to provide sufficient reason for the sake of proving her claim and winning a discussion. It represents the obligation to defend a statement “against all counter-argument or evidence” (Thayer 2019, p. 76). Hence, if a prosecutor in a criminal proceeding cannot defeat one of the defender’s counter-arguments, she fails to prove guilt beyond a reasonable doubt and, ultimately, loses the case (for details, see Strong 1992, p. 426). So, crucially, the party carrying the burden of persuasion (e.g., the prosecutor) “carries the consequences of any residual uncertainty” (Hahn & Oaksford 2007, p. 42). But fulfilling this burden in the concluding stage entails winning: proving guilt beyond a reasonable doubt in the final stage of a criminal proceeding is necessary and sufficient to win the case. The relationship between the two burdens is complex, but some differences are apparent. First, the persuasion burden is conceptually linked to the specific evidential threshold.125 While the production burden may require evidence of low and unspecified strength—evidence sufficient “to raise an issue” (Capper & Cross 1990, p. 113) or generate reasonable disagreement (Prakken & Sartor 2009, p. 244)—the persuasion burden requires evidence of high and specified strength. Second, the burden of production has procedural priority. We must first establish that all parties have produced arguments when legitimately requested, and then evaluate the strength of arguments provided by parties carrying the persuasion burden: “The burden of persuasion becomes a crucial factor only if the parties have sustained their burdens of producing evidence” (Strong 1992, p. 426). This verification order reveals the third difference. The burden of persuasion is allocated before the discussion’s beginning (Freeman 2005, p. 22) or at different points throughout a discussion (Prakken & Sartor 2009, p. 243), but whether “this burden is met is determined in the final stage of a proceeding after all the evidence is provided” (p. 243). As pragma-dialecticians may put it, while the production burden is verified at various points in the argumentation stage, the persuasion burden is verified in the concluding stage (van Eemeren & Grootendorst 2004, pp. 60-62). In summary, while the burden of production manages the introduction of arguments, the persuasion burden manages uncertainty in deliberation and becomes essential when all 124  For instance, the prosecutor can satisfy the production burden by producing reasonable evidence in favour of “John is guilty.” Still, she loses a discussion if this evidence does not meet a particular standard, e.g., if the prosecutor does not prove John’s guilt ‘beyond a reasonable doubt’ (Prakken & Sartor 2009, p. 226). 125  As Hahn & Oaksford (2007, p. 44) explain, the threshold T (e.g., beyond a reasonable doubt) is a means to translate the degree of conviction (e.g., “We are certain to degree X that John is guilty”) into a binary outcome (e.g., “Prosecutor wins” vs. “Prosecutor loses”). Since the justification of belief is a matter of degree and acting upon a belief is (typically) a yes-no matter, we need a threshold to reconcile the two domains. Accordingly, if our degree of conviction X exceeds T, we pronounce John guilty and send him to jail, and if it does not, we pronounce John innocent and set him free.

120

On Presumptions, and Asymmetrical Allocations of Burden of Proof

arguments are already introduced.126 This distinction will prove important in analysing the unique deontic function and differences between two types of presumptions. In the following sections, I argue that only practical presumptions allocate the persuasion burden, and that practical and cognitive presumptions asymmetrically allocate the production burden, but in distinct dialogical settings.

2.2 Two Meanings of Allocation There are two senses of ‘allocation.’ Allocating the burden of proof might mean that a dialogical party should defend her view (see Pinto 2001, p. 112; Scott 2005, p. 54; Prakken & Sartor 2006, p. 30; Rescher 2006, p. 21; Walton 2014, p. 51; Godden 2017, p. 489). On this interpretation, allocation determines who must produce an argument, regardless of the corresponding evidence standard, or whether the burden is assumed or assigned, incurred or shifted, generated, or indicated.127 Accordingly, the allocation where all parties must support their positions is different from the allocation where the challenger is permitted to question, or test the proponent’s view without defending her position in return. On the second interpretation, allocation of the burden of proof determines an appropriate standard (threshold) that an argument must meet to sufficiently justify some proposition. In legal studies, Kokott connects allocation to the “weight, measure, or degree of persuasion required of any party” (1998, p. 124). In argumentation theory, Walton links it to a “strength (weight) of argument required by one side to reasonably persuade the other side” (1988, p. 234). Thus, allocation varies depending on the strength of evidence needed to fulfil the burden: the allocation requiring a reason that makes a standpoint worth considering differs from the allocation requiring a reason that makes a standpoint either more plausible, or clear and convincing, or proven beyond a reasonable doubt.128

4

Two interpretations of allocation are systematically connected to general aspects of the burden of proof, expressed by the questions: (a) Who must defend a view, and (b) with 126  Also, the burden of production is often portrayed as dynamic (it shifts from one party to another), whereas the burden of persuasion is static. Moreover, the former concerns any challenged proposition, whereas the latter is typically associated only with the ultimate/initial standpoint. Finally, unlike the production burden, the persuasion burden does not seem conditional upon the opponent’s challenge. These (controversial) differences are not significant for the present analysis. 127 

That “P has the burden of proof” could mean that P assumes this obligation before a discussion’s beginning: P can

choose to defend a proposal to initiate the discussion and motivate the addressee to consider the proposal (see Kauffeld 1998, pp. 249-50). Second, the burden might be assigned to the party before a discussion’s beginning: in criminal law, the prosecutor’s burden of persuasion is allocated before the start of a trial. Third, P can incur the burden during a discussion if the interlocutor poses a challenge to P’s assertion. (For differences between incurring, assuming, and acquiring, see Tseronis 2009, Chapter 3.) Fourth, the burden can be shifted to P during a discussion if the interlocutor supports her position with sufficient argument. Furthermore, the expression “X allocates the burden on P’s side” is ambiguous because X could either generate/create or indicate/imply P’s burden. For instance, the rule ‘innocent unless proven guilty’ in criminal law generates the prosecutor’s obligation to produce reasons. By contrast, the presumptive status of a proposition “John is innocent” indicates, signals, or conceptually implies (rather than generates) the prosecutor’s burden of proof (Bodlović 2019b, p. 585). 128 

Roughly, this corresponds to four distinct standards of proof in law (see, e.g., Macagno & Walton 2012).

121

CHAPTER 4

what strength this should be done (Walton 1998, p. 246; Hahn & Oaksford 2007, p. 43; Tseronis 2009, p. 82)? To a significant extent, these meanings of allocation underlie two types of asymmetrical allocation: binary asymmetry and degree asymmetry.

3 Presumptions and Binary Asymmetries We mentioned that, supposedly, presumptions asymmetrically allocate the burden of proof. What does this asymmetry look like? If ‘allocation’ means that the party must present an argument, then, plausibly, it becomes asymmetrical when some dialogical party is not obliged to present any argument at all. I label this allocation a binary asymmetry.129 The forthcoming analysis focuses on asymmetries in mixed disputes, i.e., dialogues where parties advance contradictory standpoints (van Eemeren & Houtlosser 2002, p. 18). In such dialogues, the binary allocation means that while Proponent 1 is not obliged to present any argument for p, Proponent 2 is obliged to submit an argument for q. Binary asymmetry appears controversial because mixed disputes are, after all, symmetrical: proponents put forward positions of the same pragmatic kind (Walton 1998, p. 38).130 Depending on the strength of argument required from the burdened party, we can distinguish two binary asymmetries. First, Proponent 2 can be allowed to offer some minor argument for q. This happens when she carries the burden of production, i.e., an obligation to produce any relevant argument. I label this asymmetry a light burden binary asymmetry (LBBA). However, when Proponent 2 is obliged to provide a sufficient argument for q, presenting a minor argument is suboptimal. In this case, she carries the burden of persuasion, or the burden of successful defence (see Section 3.4). I label this asymmetry a heavy burden binary asymmetry (HBBA). LBBA and HBBA promote dialectical efficiency, but they are governed by distinct normative considerations. While LBBA is associated with the requirements of ‘staying in the game,’ HBBA is associated with rational requirements of ‘winning the game.’ Also, the distinction reveals that Proponent 1, relatively speaking, can be more or less dialectically privileged: the deontic gap between ‘not-producing’ and ‘producing sufficient argument’ (HBBA) is wider than the deontic gap between ‘not-producing’ and ‘producing some argument’ (LBBA). Distinguishing LBBA from HBBA enables us to specify central questions of Section 3. First, does presumption p entirely liberate Proponent 1 (henceforth, the presumption’s 129 

The asymmetry is binary because it operates with contradictory deontic states: one party is and another is not

obliged to produce an argument. 130  Binary asymmetry is surely applicable when parties put forward pragmatically distinct positions. In a non-mixed dispute— where the proponent asserts p, and the opponent only poses a challenge “Why p?”—burden gets allocated asymmetrically on rather trivial grounds. That is, by standard definition, the burden of proof is an obligation to defend “assertives” (van Eemeren & Houtlosser 2002, p. 17), “claims” (Freeman 2005, p. ix; Bermejo-Luque 2011, pp. 192-3), “theses” (Johnson 2000, p. 194), or “standpoints” (van Eemeren & Grootendorst 2004, p. 139) rather than doubts and challenges.

122

On Presumptions, and Asymmetrical Allocations of Burden of Proof

User) from the dialectical burden, and under what conditions? Second, if these conditions are fulfilled, is Proponent 2 (henceforth, the presumption’s Challenger) required to give an argument regardless of its strength, or should she provide an argument that meets some specific standard? By answering previous questions, we also answer the following one: Do practical and cognitive presumptions differ with respect to asymmetrical allocations of dialectical burden?

3.1 Practical Presumption and LBBA In a mixed dispute, the first kind of binary asymmetry is expressed by the following allocation rule. Light burden binary asymmetry (LBBA): If the User advances presumption p, and the Challenger advances contradictory standpoint q, then: • User-consequence: In a particular dialogical setting, the User is not obliged to present any argument for p; • Challenger-consequence: In a particular dialogical setting, the Challenger is obliged to present a relevant argument (that need not be sufficient) for q.

4

Does LBBA apply if p is a practical presumption? More pointedly, if p is an epistemically uncertain proposition, the acceptance whereof avoids harm in deliberation, should p entirely exempt the presumption’s User from dialectical burden while demanding some argument, however weak, from the Challenger? Consider a variation of the Umbrella case (Bodlović 2021). Before leaving the apartment on a winter day, Anne and Jim deliberate whether to bring an umbrella. Since they do not want to get soaked and risk becoming sick, parties agree that “If rain is uncertain, we should bring an umbrella.” Their decision, therefore, depends on rain’s certainty. After observing the cloudy sky, Anne and Jim estimate that the chance of rain is, roughly, 50%. So, at that point, we can reasonably expect them to concede “It will rain” (p) as a legitimate practical presumption.131 Nevertheless, a mixed difference of opinion arises. (t1) Anne: Presumably(p), it will rain (We should bring an umbrella). (t2) Jim: No, it will not rain (Therefore, we should not bring an umbrella).

131  The literature offers many examples of practical presumption. For instance, if a handyman is uncertain whether the ladder is safe, he should presume that it is unsafe and postpone cleaning the gutters (Dare & Kingsbury 2008; Aijaz et al. 2013). Also, if an agent is uncertain whether there is a snake on the floor or a coil of rope, he should presume that the snake is on the floor and act accordingly (Walton 2014). Furthermore, if it is uncertain whether the person who will ruin your evening will attend the party, then you should presume that she will attend it and skip the party (Bodlović 2020b). Finally, the Umbrella case is similar to the well-known loaded gun example (Walton 1988).

123

CHAPTER 4

Standpoints cancel each other out, so should Anne and Jim both present arguments in the next turn?132 First, let us investigate whether LBBA’s ‘challenger-consequence’ follows at t3. Intuitively, Jim must defend his standpoint, but can he legitimately offer a minor (insufficient) argument at t3?133 In deliberation, an argument sufficiently defeats a practical presumption only by meeting a specified threshold (proof standard). The threshold’s allocation depends on various contextual factors, but primarily on values and preferences. In the Umbrella case, Anne and Jim share a hierarchy of values: both value health over convenience, and, consequently, prefer not to get soaked and risk sickness. If their shared preference is stronger, the threshold for successfully refuting a presumption is set higher—e.g., “It will not rain” must be at least 80% probable to leave the umbrella at home. If their shared preference is weaker, the threshold is lower—e.g., “It will not rain” must be at least 60% probable to persuade Anne to leave the umbrella. What if Jim gives insufficient argument? Suppose that parties set the threshold at 60%, and he says: (t3) Jim: The weather forecast estimates “It will not rain” is 55% probable. Although the argument is not persuasive, it seems permissible in the sense of not being fallacious: it does not undermine the basic rationality of deliberation. However, one may wonder if Jim’s argument achieves anything. Since it cannot defeat a practical presumption and change a default course of action, isn’t it entirely superfluous? How is providing a minor argument better than giving no argument at all? Simply put, offering a minor argument enables the Challenger to ‘stay in the game.’ Without arguing at t3, Jim would violate the obligation-to-defend rule by evading or shifting the burden of proof (van Eemeren & Houtlosser 2015, p. 170) and lose the discussion on a controversial statement. In contrast, by submitting a minor argument, he creates conditions for the further development of deliberation. Since agents share preferences in the Umbrella case, Anne can consider Jim’s minor argument in a cooperative fashion by responding, for instance: 132 

If they should, then who presents an argument at t3 is the matter of who goes first, rather than genuine asymmetry

in the burden of proof. Also, a background assumption is that neither Anne nor Jim is willing to retract the standpoint once it gets rejected. 133  In my opinion, Jim incurs the burden at t3, although Anne does not explicitly demand argument by that time. This is because, by advancing “It will not rain,” Jim refuses to concede a position he is reasonably expected to concede. Namely, he initially accepts “If the sky is cloudy, the rain is 50% probable,” and the sky is cloudy at t2, so Anne (as well as some external analyst) would expect him to remain committed to “The rain is 50% probable.” Since Jim keeps track of his previous commitments, pays attention to publicly available evidence, and has a basic theory of mind, he must be aware that advancing “It will not rain” at t2 seems controversial, if not inconsistent. As a reasonable communicator, Jim owes Anne some explanation and, as a reasonable arguer, he should know that only stating “It will not rain” will not change Anne’s mind (given the cloudy sky). So, arguing for his standpoint seems like an obvious move that comes naturally in these particular circumstances. A reasonable agent will (and should) offer argument automatically, almost by some dialectical reflex, because, in some sense, Anne’s challenge is contextually implied. At least in typical circumstances, it can be anticipated as soon as Jim advances “It will not rain.”

124

On Presumptions, and Asymmetrical Allocations of Burden of Proof

(t4) Anne: 55% is almost persuasive. To be sure we are making the right decision, check some alternative source. This gives Jim another chance to defend his standpoint. Suppose he checks an alternative weather forecast and learns that “It will not rain” is 65% probable. In this case, his minor argument from t3 represents a legitimate, intermediate step towards choosing the right course of action. Thus, at t3, Jim carries the burden of production (see Section 2.1)—to avoid unsupported claims, and doing what is necessary to win in the deliberation’s final stage, he is obliged to provide an argument, and permitted to give a minor argument for “It will not rain.” Accordingly, in the Umbrella case, practical presumption entails LBBA’s ‘challenger-consequence.’ What about the ‘user-consequence’? Is Anne exempted from defending “Presumably(p), it will rain” immediately after advancing it? Intuitively, she is. That is, Anne’s rationale is transparent at t3.134 Since parties previously agreed that rain is uncertain and getting soaked more harmful than (needlessly) carrying an umbrella, Jim should know that Anne considers bringing an umbrella on these particular grounds. Since repetition is not obligatory, Anne is not committed to argue at t3. Also, note that Anne may challenge Jim’s standpoint at t3.

4

(t1) Anne: Presumably(p), it will rain. (t2) Jim: No, it will not rain. (t3) Anne: Why do you think it will not rain? From a dialectical perspective, challenging appears more desirable than defending: by requesting the non-transparent reasons for Jim’s unexpected standpoint, Anne contributes more to deliberation than by repeating transparent reasons for her expected standpoint. Since challenging is permitted, and even recommended at t3, Anne is not obliged to present an argument. But is she exempted from burden at t3 only because Jim has not challenged her standpoint yet? If Jim requests an argument at t3, should Anne present it at t4?135 This does not seem plausible. (t1) Anne: Presumably(p), it will rain. (t2) Jim: No, it will not rain. (t3) Jim: Why do you presume that it will rain? 134 

While analysing entitlements and obligations of Proponent 1, I assume that t3 was not already ‘occupied’ by the

Challenger and vice versa. Throughout the analysis, as a matter of convention, I take the turn t3 as representing the first turn available to some party immediately after advancing a standpoint. 135  Some theorists portray the burden of proof as a conditional obligation only arising if the disagreeing party explicitly demands proof (Kauffeld 1998, p. 263; van Eemeren & Houtlosser 2002, pp. 17-18; van Eemeren & Grootendorst 2004, p. 139; Tseronis 2009, p. 92; van Laar & Krabbe 2013, p. 202).

125

CHAPTER 4

Given the presumption’s transparent justification, Jim’s challenge is unnatural and hardly contributes to reasonable dialogue. How, exactly, does a request to repeat an already known justification help agents to “[d]ecide [on the] best available course of action” (Walton 2006, p. 183)? Jim’s request at t3 is unproductive, or even counterproductive, for it wastes time when efficient time-management is essential, i.e., when parties must choose the right action before the deadline.136 Since “commonsense guide to where a speaker has a burden of proof is whether proof can fairly be demanded of the advocate” (Kauffeld 1998, p. 263), Anne does not have the burden at t4.137 So, practical presumption entails the ‘user-consequence’ in two dialogical settings. First, when Jim advances a contradictory standpoint at t2, Anne is not obliged to argue at t3. Second, when Jim requests justification at t3, Anne is not obliged to argue at t4. The next question is: If Jim gives an argument at t3, does Anne incur the burden at t4? This depends on the argument’s strength. Suppose Jim offers a minor argument at t3, such as “The weather forecast estimates ‘It will not rain’ is 55% probable.” In this case, Anne is permitted, but not obliged to argue at t4. She may demand further support, and reply: “55% is almost persuasive. To be sure we are making the right decision, check some alternative source.” But what if Jim offers a sufficient argument at t3? (t1) Anne: Presumably(p), it will rain. (t2) Jim: No, it will not rain. (t3) Jim: The weather forecast estimates “It will not rain” is 75% probable. In this case, Anne must provide an argument at t4 because, otherwise, she (probably) loses the discussion.138 Since our analysis centres on presumptions and burdens of proof, two questions become essential. 136 

Typically, practical presumptions are used in response to the so-called ‘deliberation problem,’ consisting of three

components: (1) a deliberation dilemma (To bring the umbrella or not?), (2) evidential uncertainty (Rain is uncertain), and (3) deliberation pressure (Parties must decide soon, otherwise they will be late) (Bodlović 2021). Deliberation pressure makes efficient time-management essential, thereby rendering some dialectical moves more desirable than others. For instance, arguing is more desirable than explaining because it is more likely to influence deliberation. Accordingly, explaining wastes time that could be spent more constructively (for details, see Bodlović 2020b, pp. 284-287). 137  As van Eemeren & Houtlosser observe, “maintaining the burden of proof does not make sense when the protagonist has earlier defended his standpoint successfully against the same antagonist starting from the same point of departure” (2002, p. 18). 138 

Suppose that, at t4, Anne says “Your argument is sufficient, but I am not convinced,” or “Your argument appears

sufficient, but it is not.” Without further justification, both responses seem deviant. This becomes clear when analysed from Jim’s dialogical perspective. On the one hand, since reasonable dialogue is hardly possible when somebody openly ignores sufficient arguments, he seems allowed to terminate a discussion. On the other hand, he is undoubtedly entitled to challenge Anne’s response by asking, “Why are you not convinced by sufficient argument?” or “Why does my argument only seem sufficient?” However, Anne’s non-argumentative response is deviant under two conditions. First, she must understand all relevant details about Jim’s reasoning from t3. Otherwise, she would be permitted to ask for clarification, such as “Which weather forecast are you referring to, exactly?” The second condition is that she wants to (rationally) retain her standpoint at t4.

126

On Presumptions, and Asymmetrical Allocations of Burden of Proof

First, does Anne incur the burden of proof at t4, or some other kind of burden of arguing? On the standard definition, the burden of proof denotes an obligation to support or defend one’s view, but Anne is also allowed to attack Jim’s argument at t4.139 She might reply: (t4) Anne: Jim, the weather forecast website you are referring to is highly unreliable. This is because… From a pragmatic perspective—that studies arguments as tools used in various ways, by different agents, and for many purposes—arguing to attack someone’s opinion differs from arguing to defend one’s position (Walton 1996b, pp. 3-4). Also, from the perspective of justificatory relations, defeating Jim’s argument does not support “It will rain,” i.e., the proposition directly connected to defending Anne’s proposal “We should bring an umbrella.” Rather, Anne’s undercutting defeater justifies adopting an agnostic stance about the rain’s probability, insofar as the particular website is concerned (see Pollock 1987, p. 485). Since Anne does not defend her position at t4, does she discharge the burden of proof? I believe she does. Namely, after Anne undercuts Jim’s argument, two pieces of evidence become available: the initial estimation that rain is 50% probable, and the unreliable website’s estimation. On these grounds, Anne and Jim should agree the rain’s probability is around 50%. Since the successful defence of “Presumably(p), it will rain” requires that rain’s probability is over 40%,140 an undercutter makes Anne’s standpoint acceptable. So, despite pragmatic differences, attacking and defending are dialectically connected—by attacking Jim’s argument, ultimately, Anne defends her standpoint. Also, note that pragmatic differences become elusive when the attack is a part of the larger, strategically-planned defence. To conclude, Anne carries the burden of production at t4, but she could (also) discharge it by refuting Jim’s argument.

4

Since Anne incurs the burden of proof at t4, LBBA’s ‘user-consequence’ does not follow. But to understand the relationship between practical presumptions and binary asymmetries, we must answer the second essential question: Does Anne become obliged to defend a presumption, or just an ordinary statement at t4? It may seem that Anne’s burden reveals nothing about presumptions and binary asymmetries. Since sufficient arguments defeat presumptions,141 and Jim offers sufficient 139 

The burden of proof is portrayed as an obligation “to support one’s view” (Aijaz et al. 2013, p. 260), “furnish

support for the thesis” (Johnson 2000, p. 194), or “provide evidence for your position” (Dare & Kingsbury 2008, p. 503). Also, it is characterised as an obligation to “defend, by argument, certain claims that we make” (Govier 2010, p. 166), “defend the standpoint once challenged to do so” (Eemeren and Houtlosser 2002, p. 17), or “defend any premise that the audience will not be able to accept without support” (Johnson 2000, p. 194). It is explained in terms of a “defence norm” (Rescorla 2009a, p. 88) and the “obligation-to-defend rule” (van Eemeren & Grootendorst 2004, p. 191). 140  Because, in this example, the agreed standard of proof for the contrary standpoint “It will not rain” (and thereby for defending “We should not bring an umbrella”) is 60%. 141  Namely, “presumptions can be undercut or defeated” (Freeman 2005, p. x), and a presumption “stands good unless and until its good standing is disproved” (Lewiński 2017, p. 592). Accordingly, p has presumptive status until it is “defeated by case-specific considerations” (Rescher 2007, p. 30), i.e., until “new evidence comes in showing that it is false” (Walton

127

CHAPTER 4

argument against “It will rain” at t3, plausibly, Anne is not obliged to support a practical presumption at t4. She still remains exempted from the burden at every stage where p is a practical presumption. This straightforward argument, however, rests on a narrow, local conception of practical presumption. Admittedly, “according to standard accounts, presumptions are dialectically privileged, yet defeasible, propositions” (Bodlović 2020b, p. 289). But although we must consider presumption’s defeasibility, we should also consider the nature of its dialectical privilege. Regardless of Jim’s defeat at t3, practical presumption p remains privileged in one important sense: p’s successful defence requires meeting a lower standard of proof than q’s successful defence. Crucially, this privilege is global: throughout the whole discussion, due to the policy of avoiding greater harm and asymmetrical allocation of the persuasion burden (see Section 2.1), Anne’s standard of proof is lighter than Jim’s.142 In this particular sense, Anne’s standpoint p represents a global practical presumption even after Jim successfully defeats p at t3.143 Hence, practical presumption p does not entail the ‘user-consequence’ in the third dialogical setting: if the Challenger sufficiently defeats (local) presumption p at tn, the User must support (global) presumption p at tn+1. The User must fully accept the proponent’s role at tn+1, and some “internal dialectical shift” occurs (see Krabbe 1995, p. 256). We might describe this shift in terms of asymmetrical allocations before, and after tn+1. Before tn+1, p entailed binary asymmetry: the User (Anne) was not obliged to argue because p was a local presumption. From tn+1, p entails a degree asymmetry; the User is obliged to argue because the Challenger (Jim) defeated p’s status of local presumption. But since p remains a global presumption at tn+1, the argument for p is assessed by a lower standard of proof than the argument for q.144 I discuss degree asymmetries in Section 4.

2014, p. 117) and “some sufficient reason is adduced against it” (Whately 1963, p. 112). This coheres with pre-theoretical intuitions since “[t]he ordinary-language word ‘presumption’ is suggestive of something tentative, contestible, reversible” (Ullmann-Margalit 1983, p. 149). 142  As we have seen, Anne must show that rain is over 40% probable in order to defend “We should bring an umbrella,” while Jim must show that “It will not rain” is at least 60% probable to support “We should not bring an umbrella.” 143 

Notably, the global practical presumption remains defeasible: if Jim ultimately satisfies his persuasion burden, p is

defeated at the concluding stage. 144  That p ceases to be local but remains a global presumption at tn+1 reveals something about the presumption’s formal definition. That is, Freeman remarks that “[t]o speak of a statement in itself as … having the status of a presumption is to speak elliptically” since “[a] statement is always a presumption relative to some juncture in some interchange” (2005, p. 26). Accordingly, he defines presumption as a relation consisting of a statement, a point in a dialectical exchange, and a person’s perspective in a discussion (e.g., proponent, challenger, or external adjudicator) (pp. 26-33). Our analysis reveals that even if all elements of a relation have fixed values in Freeman’s model, a presumptive status can be ambiguous. In the Umbrella case, from Anne’s perspective, in turn t4, p is not a local presumption in the sense of ‘freeing Anne from the burden of production,’ but is a global presumption in the sense of ‘freeing Anne from the burden of persuasion.’ So, if the presumption is a relation, a sophisticated theory should incorporate a connection between a relevant statement and distinct kinds of burdens of proof.

128

On Presumptions, and Asymmetrical Allocations of Burden of Proof

This concludes our analysis of conditions under which practical presumption p should entail both consequences of LBBA rule: A. Entailing LBBA’s ‘user-consequence’ The User is not obliged to present any argument for practical presumption p if: 1. Initially, the justification of p’s presumptive status is transparent. In this case, the User (Anne) does not carry the production burden either at t3 (after the Challenger advances q) or at t4 (after the Challenger introduces a challenge “Why ‘Presumably(p), p’?”). 2. Further in the argumentation stage, the Challenger (Jim) produces a minor argument against p. If the Challenger produces sufficient argument against p at tn, the User must support a presumption at tn+1 (incurs the production burden): binary asymmetry shifts to degree asymmetry. B. Entailing LBBA’s ‘challenger-consequence’ The Challenger is obliged to present an argument of whatever strength for q if:

4

3. We limit our analysis solely to the production burden. In this case, the Challenger (Jim) is permitted to provide a minor argument for q somewhere in the argumentation stage (e.g., at t3). However, the ‘challengerconsequence’ does not follow if we also consider the allocation of the persuasion burden (evaluated in the concluding stage).

3.2 Cognitive Presumption and LBBA Our next question is: Does LBBA apply if p is a cognitive presumption? If p is the most plausible truth-candidate the acceptance whereof promotes the goal of epistemic inquiry, should p entirely exempt the User from dialectical burden while demanding some argument, however weak, from the Challenger? Consider the Directions case where two tourists want to visit the city center (Bodlović 2019b). Jill, a local person, suggests they take the second street to the right. Since they find her testimony convincing, Smith and Jones accept “This street leads to the city center” (p) as the most plausible proposition: they concede p as an initial cognitive presumption.145 However, after a short walk, disagreement arises.

145 

The literature offers many similar examples of cognitive presumption. For instance, in the Flat-Earth case, Diane

should presume that the Earth is globe-shaped (based on a reliable source, such as the scientific consensus), and a flatEarther Steve should carry the burden of proof (Aijaz et al. 2013; Bodlović 2020b). Also, in the Cat case, if Andy sees that his cat in the tree, he should presume that his cat is indeed in the tree unless there is a sufficient reason to think otherwise (Bodlović 2021).

129

CHAPTER 4

(t1) Smith: I am happy that we are nearly at the city center! (t2) Jones: This street does not lead to the city center. Intuitively, Jones incurs the burden of proof at t3 because he suddenly retracts his previous, reasonable commitment. But is giving a minor argument permissible, or even dialectically useful? Suppose Jones provides a limited defence by saying, e.g.: (t3) Jones: This street is getting dirtier and dirtier.146 Jones’ argument is permissible because it does not instantiate a reasoning pattern systematically undermining the goals of producing knowledge (Rescher 2007, p. 37) or proving a hypothesis (Walton 2006, p. 183). Also, it potentially contributes to sufficient justification since minor arguments can gain “cumulative force” (Koplin & Selgelid 2015, p. 600). When assessed in isolation, “This street is getting dirtier” cannot justify “This street does not lead…” but suppose Jones also introduces other reasons: (t3) Jones: This street is getting dirtier and dirtier. It goes uphill, and the centre is in the valley. The traffic is lighter than before, and we have just passed an industrial warehouse. Since these considerations indicate Smith and Jones are heading towards the suburbs, they belong to the same “cluster” (Krabbe & van Laar 2021) and can be evaluated together. As soon as this is done, minor arguments become parts of one complex, more substantial argument, and provide “convergent support” (Govier 2010, p. 55) strong enough to defeat Smith’s presumption. Hence, not only are they permissible, but also dialectically effective. Therefore, Jones carries the burden of production (see Section 2.1); to avoid unsupported claims, and make tentative steps towards a successful defence, he is obliged to argue at t3, and allowed to support “This street does not lead…” with minor argument. In the Direction’s case, cognitive presumption entails LBBA’s ‘challenger-consequence.’ What about the ‘user-consequence?’ Is Smith exempted from defending “Presumably(c), this street leads to the city center?” Let us consider three dialogical settings. First, after Jones advances a contradictory standpoint at t2, Smith is not obliged to argue at t3. His rationale is transparent, and covering old ground is an ineffective epistemic strategy (see Aijaz et al. 2013, p. 273). Second, after Jones demands an argument at t3, Smith is exempted from burden at t4: due to the transparency of Smith’s justification, “Why do you think this street leads…?” is a deviant challenge. However, in the third setting, Jones gives an argument. If his argument is insufficient, like “This street is getting dirtier,” Smith does not incur the burden at t4.147 146 

Usually, suburban streets are dirtier than streets near the city center, but Jones’ justification neither matches nor

exceeds the strength of Jill’s testimony. Since locals always know the way to the city center, and Jill’s testimony was not suspicious, “This street leads to the city center” remains more plausible than Jones’ alternative. So, Jones’ minor argument is insufficient to defeat cognitive presumption p. 147 

130

For instance, he is allowed to demand additional justification by saying: “That the street is getting dirtier is weird,

On Presumptions, and Asymmetrical Allocations of Burden of Proof

However, if the argument is strong (like “This street is getting dirtier, and goes uphill; the traffic is lighter, and we have just passed an industrial warehouse”), Smith must accept Jones’ position or support his own view. For current analysis, the following is crucial: after Jones’ sufficient defeat at t3, “This street leads…” ceases to be the most plausible proposition, thereby losing the status of cognitive presumption. Smith’s burden at t4, therefore, is irrelevant to studying cognitive presumptions and asymmetrical allocations. Once p becomes defeated, cognitive presumption p, unlike practical presumption p, entails the ‘user-consequence.’ This shows that cognitive and practical presumptions free their proponents from production burden under different conditions. Practical presumption p entails the ‘userconsequence’ only if the Challenger offers a minor defeater. Namely, if Jim sufficiently defeats p at tn, two things happen: “It will rain” remains a global practical presumption, and Anne must support it at tn+1. So, if p is sufficiently defeated at tn, practical presumption p does not entail the ‘user-consequence’ at tn+1. By contrast, in the Directions case, Jones’ sufficient argument at tn defeats entirely p’s presumptive status; at tn+1, “This street leads…” is not a cognitive presumption of any kind.148 In effect, Smith’s burden at tn+1 is consistent with the hypothesis that cognitive presumptions entail LBBA. Regardless of whether the Challenger offers a minor, or sufficient defeater, cognitive presumption p entails the ‘userconsequence.’

4

This concludes our investigations of whether the deontic function of cognitive presumption entails light burden binary asymmetry. It does, under the following conditions: A. Entailing LBBA’s ‘user-consequence’ The User is not obliged to present any argument for cognitive presumption p if: 1. Initially, the justification of p’s presumptive status is transparent. In this case, the User (Smith) does not carry the production burden either at t3 (after the Challenger advances q) or at t4 (after the Challenger introduces a challenge “Why ‘Presumably(c), p’?”). If the Challenger (Jones) produces sufficient argument against p at tn, then the User incurs the production burden but he is not obliged to defend a presumption at tn+1.

but it does not prove we are going in the wrong direction. Do you have other reasons in mind?” 148  The concept of global presumption, in a sense explained in Sect. 3.1, does not apply to epistemic contexts. Namely, it rests on the differential allocation of proof standards, and it is at best controversial whether different proof standards promote or undermine epistemic goals (see Sect. 4.1).

131

CHAPTER 4

B. Entailing LBBA’s ‘challenger-consequence’ The Challenger is obliged to present an argument of whatever strength for q if: 2. We limit our analysis solely to the production burden. In this case, the Challenger (Jones) is permitted to provide a minor argument for q somewhere in the argumentation stage (e.g., at t3). However, the ‘challenger-consequence’ does not follow if we also consider the allocation of the ‘burden of successful defence’ in the concluding stage (see Section 3.2)

3.3 Practical Presumption and HBBA The following allocation rule expresses the second kind of binary asymmetry. Heavy burden binary asymmetry (HBBA): If the User advances presumption p, and the Challenger advances contradictory standpoint q, then: • User-consequence: In a particular dialogical setting, the User is not obliged to present any argument for p; • Challenger-consequence: In a particular dialogical setting, the Challenger is obliged to present a sufficient argument for q (assessed in the concluding stage). Does HBBA apply if p is a practical presumption? In the Umbrella case, Jim must support “It will not rain” with any relevant argument at t3, but should he present a sufficient argument at some later point? He should. The Umbrella case resembles a criminal trial in several respects. First, Anne and Jim must decide whether to proceed on “It will (not) rain” before the deadline. Second, “It will (not) rain” may remain uncertain when the deadline arrives. Third, when uncertain, accepting one particular standpoint seems pragmatically desirable: proceeding on “It will rain” avoids more significant harm. As a result, Jim’s argument must meet a demanding standard of proof to justify “It will not rain” in the concluding stage (e.g., “It will not rain” must be at least 60% probable). Just like the prosecutor in the criminal case, Jim carries “the risk that inconclusive findings will count against [him]” (Scott 2005, p. 54): by failing to produce sufficient argument, he loses, and Anne wins. So, Jim carries the burden of persuasion (see Section 2.1): he is obliged to support “It will not rain” with sufficient argument because, otherwise, he will lose the debate in the concluding stage. In the Umbrella case, practical presumption entails HBBA’s ‘challengerconsequence.’ What about the ‘user-consequence?’ Is Anne exempted from defending “Presumably(p), it will rain?” In Section 3.1, we explained that Anne should argue at tn+1 only if Jim offers a strong argument at tn. However, by incurring the production burden at tn+1, Anne does 132

On Presumptions, and Asymmetrical Allocations of Burden of Proof

not incur the burden of persuasion. To see this, suppose rain remains, for instance, 50% probable when the deadline arrives. Due to precaution, agents should proceed as if it will rain: Anne’s standpoint gets accepted, and she wins. In contrast, Jim fails to produce an argument that makes “It will not rain” at least 60% probable, and loses the discussion. So, only Jim carries the “risk of non-persuasion” (Wigmore 1962, p. 283), and Anne’s standpoint represents a dialectically privileged, global presumption. In sum, practical presumption p entails heavy burden binary asymmetry under the following conditions: A. Entailing HBBA’s ‘user-consequence’ The User is not obliged to present any argument for practical presumption p if: 1. Initially, the justification of p’s presumptive status is transparent. In this case, the User (Anne) does not carry the production burden either at t3 (after the Challenger advances q) or at t4 (after the Challenger introduces a challenge “Why ‘Presumably(p), p’?”). 2. Further in the argumentation stage, the Challenger (Jim) produces a minor argument against p. If the Challenger produces sufficient argument against p at tn, the User must support a presumption at tn+1 (incurs the production burden): binary asymmetry shifts to degree asymmetry.

4

B. Entailing HBBA’s ‘challenger-consequence’ The Challenger is obliged to present a sufficient argument for q if: 3. We limit our analysis solely to the persuasion burden (assessed in the concluding stage). However, the ‘challenger-consequence’ does not follow if we also consider the allocation of the production burden in the argumentation stage. Importantly, LBBA and HBBA are not rival rules of asymmetrical allocation. Treating them separately is useful and adequate, but since LBBA and HBBA are related to distinct kinds of the burden of proof, typically, they co-exist in a reasonable discussion.

3.4 Cognitive Presumption and HBBA Does HBBA apply if p is a cognitive presumption? In the Directions case, Jones carries the production burden at t3, but should he successfully defend “This street does not lead to the city center” at some point? He should. Namely, Jones’ ultimate goal is winning the discussion, i.e., persuading Smith that “This street does not lead…” is justified/true belief. To achieve this goal, Jones

133

CHAPTER 4

must present an argument providing stronger justification than Jill’s testimony.149 He can offer a complex argument—such as “This street is getting dirtier, and goes uphill; the traffic is lighter, and we have just passed an industrial warehouse”—or a simple, but decisive one: (t1) Smith: I am happy that we are nearly at the city center! (t2) Jones: This street does not lead to the city center. (t3) Jones: Look, the town sign is crossed! So, in addition to elementary rules of critical discussion, Jones must follow the optimality rule “Try to win” (see Krabbe 2001, p. 6), or “[T]ry to develop a conclusive defence” (Krabbe & van Laar 2011, p. 222). He carries both the production burden, and the burden of a successful defence. For the analysis of presumptions, the following is crucial: the burdens of persuasion and successful defence are not identical. The persuasion burden entails that if Jim’s standpoint (ultimately) remains uncertain, the case is decided against him, i.e., Anne wins. However, sometimes the burden of successful defence entails a different result in similar circumstances. In the Directions case, suppose Jones’ argument is equally strong as Jill’s testimony. This surely renders Jones’ standpoint “This street does not lead…” uncertain, but the case is not decided against him. Instead, reasonable parties should suspend judgment, and agree that nobody wins. In such conditions, treating Smith as the winner and “This street leads…” as a justified/true belief is epistemically unreasonable. As Cargile puts it: The consequence of having not-p assumed against you is only one possible penalty for failure to meet the burden of proof. Another is the much milder consequence of not being able to assume p in your argument. (Cargile 1997, p. 60) Thus, although burdens of persuasion and successful defence are concerned with winning and evaluated in the concluding stage, they entail distinct conditions of losing—only those having the persuasion burden carry “the consequences of any residual uncertainty” (Hahn & Oaksford 2007, p. 42). They are distinct dialectical burdens, applicable to different contexts, and related to different types of presumptions.150 149 

If Jones’ argument is equally strong as Jill’s testimony, nobody wins. The argument is insufficient to defend Jones’

standpoint q (“This street does not lead…”) but sufficient to defeat a presumptive status of p—Smith becomes obliged to suspend judgement, or defend p. If Jones’ argument is weaker than testimony, Smith wins, and Jones loses. 150 

Also, burdens presuppose different proof standards. Typically, the proof standard associated with the persuasion

burden includes an exact threshold (e.g., “It will not rain” must be at least 60% probable), determined before deliberation, in connection to values/preferences and pragmatic policies (e.g., avoiding harm, reducing risk, etc.). By contrast, the standard associated with the burden of successful defence is relative to arguments produced in the dialogue at hand. Typically, the standpoint is not supposed to meet some pre-determined degree of plausibility. Instead, to get accepted, it must be more plausible than the alternative standpoint. That is, according to the basic rule of dialectics, “in a context of contentions the stronger arguments prevail” (Rescher 2006, p. 10). For a detailed analysis of various limitations relevant to applying the persuasion burden in epistemic (persuasion) dialogues, see Hahn & Oaksford (2007).

134

On Presumptions, and Asymmetrical Allocations of Burden of Proof

Accordingly, in the Directions case, the cognitive presumption entails HBBA’s ‘challenger-consequence:’ Jones is obliged to construct a sufficient argument before the concluding stage. What about the ‘user-consequence?’ Is Smith obliged to argue for “Presumably(c), this street leads to the city center” at some point? In Section 3.2, we explained Smith should argue at tn+1 only if Jones offers a strong argument at tn. However, by incurring the production burden at tn+1, Smith does not incur the burden to support a cognitive presumption: since p’s presumptive status is defeated, the ‘user-consequence’ follows as long as p is a presumption. But what about the burden of successful defence? Is Smith obliged to provide sufficient argument before the concluding stage? I believe he is not. Smith can contribute by requesting clarifications, explanations, or justifications, and is also permitted to attack Jones’ position with arguments (both weak and strong ones). But as long as p remains the most plausible proposition on transparent grounds, Smith is not obliged to provide sufficient argument for p. In sum, cognitive presumption p entails heavy burden binary asymmetry under the following conditions: A. Entailing HBBA’s ‘user-consequence’

4

The User is not obliged to present any argument for cognitive presumption p if: 1. Initially, the justification of p’s presumptive status is transparent. In this case, the User (Smith) does not carry the production burden either at t3 (after the Challenger advances q at t2) or at t4 (after the Challenger poses a challenge “Why ‘Presumably(c), p’?” at t3). If the Challenger (Jones) produces sufficient argument against p (for q) at tn, then the User (Smith) incurs the production burden at tn+1 but he is not obliged to defend a presumption. B. Entailing HBBA’s ‘challenger-consequence’ The Challenger is obliged to present a sufficient argument for q if: 2. We limit our analysis solely to the allocations of the burden of successful defence in the concluding stage. However, the ‘challenger-consequence’ does not follow if we only consider the allocation of the production burden in the argumentation stage. Although Smith is exempted from providing sufficient argument, he is undoubtedly obliged to possess one. Since he cannot win the discussion without having sufficient argument accessible to Jones, cognitive presumption allocates the attitudinal burden symmetrically (see Section 2.1). Importantly, in epistemic contexts, attitudinal burden commits parties to an identical standard of proof. When evaluated in the concluding stage, standpoints are treated equally: Smith’s standpoint “This street leads to…” should not

135

CHAPTER 4

be accepted as (epistemically) justified if, ultimately, Smith has a weaker argument than Jones. Since things work differently in the Umbrella case, the notion of attitudinal burden explains another difference between practical and cognitive presumptions: only practical presumptions asymmetrically allocate the obligation to possess sufficient argument.151 The previous results are theoretically puzzling. Note that the User is exempted from dialectical burden due to the transparency of the foundations of presumption. But justification’s transparency is contingent: it is neither a necessary nor sufficient feature of a presumption. Proposition p may be a presumption (in an objective sense, external to dialogue at hand) even if the Challenger does not find p’s justification transparent (assuming that p is either the most plausible or pragmatically desirable proposition). Also, p may not be a presumption even when p’s justification is transparent since many controversial positions have transparent justifications.152 This raises the following questions: How can the burden’s asymmetrical allocation, traditionally understood as presumption’s defining dialectical feature (Godden & Walton 2007, p. 334), be so closely associated with presumption’s contingent feature? And if controversial assertions may also exempt proponents from dialectical burdens, why should we consider presumption’s deontic function unique? I leave these questions for another occasion. But the previous analysis helps us to distinguish between several normative aspects that explain (binary) asymmetrical allocations of the dialectical burden. The Challenger may be burdened, and the User exempted from burden due to strength, adequacy, legitimacy, or transparency (accessibility) of the foundations of presumption. 1. Strength: the foundations of presumption make p the most plausible or pragmatically desirable proposition. Usually, this objective condition is external to the discussion at hand. 2. Adequacy: the foundations of presumption are acceptable to the Challenger. This dialectical (rhetorical) condition is internal to a discussion at hand. 3. Legitimacy: the foundations of presumption are widely accepted, i.e., acceptable to the Challenger’s peers. This represents a broader social condition. 4. Transparency/accessibility: the foundations of presumption are apparent, wellknown, or easily reconstructable.

151 

For further discussion about how practical and epistemic contexts entail different allocations of burdens of proof,

either dialectical or attitudinal, see Hahn & Oaksford (2007), Dare & Kingsbury (2008), Koplin and Selgelid (2015), and especially Aijaz et al. (2013). 152 

Suppose discussants are familiar with the particular conspiracy theory underlying “The Earth is disc-shaped.” In

that case, flat-Earther Steve may be exempted from providing reasons although his standpoint is not a presumption.

136

On Presumptions, and Asymmetrical Allocations of Burden of Proof

4 Presumptions and Degree Asymmetries In every asymmetrical allocation, “[o]ne person has a more demanding job of proving a point” (Allen 2004, p. 81). The Challenger’s probative job is more demanding in binary asymmetries because the User carries a “weightless burden” (Aijaz et al. 2013, p. 264), but the Challenger’s burden might be heavier even when the User must produce some argument. The allocation where both discussants produce arguments, but the Challenger is burdened to a greater degree, I label degree asymmetry. Section 4 delineates different kinds of degree asymmetries, and analyses their relations to practical and cognitive presumptions. It is important to stress that, typically, rules concerning degrees asymmetries are not rival to those concerning binary asymmetries, for they correspond to different dialogical circumstances (e.g., circumstances where the User’s justification is not entirely transparent). Also, by recognising that the required strength of an argument is not the only factor determining the Challenger’s burden, degree asymmetries introduce additional considerations into dialectical models. For instance, even when the User and the Challenger must meet an identical evidential standard, the Challenger may carry a heavier burden if the (strong) evidence supporting her view is not easily accessible.

4

4.1 Assessment Criteria Degree Asymmetries The Challenger might carry a heavier burden because, ultimately, her argument should meet more rigorous criteria to successfully support the standpoint than the User’s argument. The first, paradigmatic type of assessment criteria degree asymmetry is linked to asymmetrical allocation of the standard of proof. Proof standard degree asymmetry (PSDA): If the User advances presumption p, and the Challenger advances contradictory standpoint q, then: • User-consequence: In a particular dialogical setting, the User is obliged to present an argument for p that meets the proof standard S; • Challenger-consequence: In a particular dialogical setting, the Challenger is obliged to present an argument for q that meets the higher proof standard S* (assessed in the concluding stage). If p is a presumption with transparent rationale, p does not entail PSDA’s ‘userconsequence’ (as shown in Section 3). However, the User may become obliged to support presumption p once p’s rationale is not entirely clear. Suppose that Mark and Alice are deliberating whether to attend a friend’s party (see Bodlović 2020b). Their decision depends on John’s whereabouts, and, since both Mark and Alice intensely dislike John, they accept the following concluding rule: “Ultimately, if there is a reasonable chance John might attend the party, we shall proceed on ‘John will attend the party,’ and skip it.” But although “John will attend the party” (p) becomes a practical presumption, parties disagree on how to proceed. 137

CHAPTER 4

(t1) Mark: Presumably(p), John will attend the party (We should skip it). (t2) Alice: We should attend the party. Crucially, Alice is allowed to challenge Mark’s practical presumption at t3 by saying: (t3) Alice: The precaution is fine with me. But what makes you think John might attend the party? Namely, at that time, “John might attend the party” is not obviously plausible: Mark and Alice have not reached an explicit agreement about it previously and, supposedly, cannot rely on immediately accessible evidence regarding John’s whereabouts.153 So, since Mark’s justification is unclear to Alice, her challenge is perfectly legitimate, and Mark incurs the burden of production at t4. Unlike Alice, he is not committed to the higher proof standard in the concluding stage, but he is obliged to present some (minor) arguments throughout the argumentation stage. So, first, practical presumption p entails PSDA’s ‘user-consequence’ if the justification of presumptive status is not transparent. Second, the User must defend practical presumption at tn+1 if the Challenger sufficiently defeats p at tn (as argued in Section 3.1). Finally, practical presumption entails ‘challenger-consequence:’ due to pragmatic considerations (safety), the Challenger is ultimately committed to the higher standard of proof (see Section 3.1 and Section 3.3). Does cognitive presumption entail PSDA? In the Flat Earth example, Diane and Steve discuss Earth’s shape (Bodlović 2020b). At t1, Diane asserts a strong cognitive presumption and, at t2, Steve rejects it. (t1) Diane: The Earth is globe-shaped (p) (t2) Steve: No, the Earth is disc-shaped (q). The question is: If Diane’s justification is not transparent to Steve, and she must defend her standpoint, is her argument, ultimately, assessed by a lower standard of proof? In other words, are eccentric individuals committed to a higher standard to epistemically justify their controversial standpoints? Recently, several scholars have argued that, in epistemic contexts, we should not allocate proof standards asymmetrically. Dare & Kingsbury (2008) maintain that in “truthdirected practices, it is normally improper to place a higher burden of proof on one side of a debate” (p. 513) because PSDA has problematic epistemic consequences. For instance, in the well-known H. pylori affair, scientists have accepted the weaker hypothesis “Gastric acid causes ulcers” for several years only because the alternative hypothesis “Bacteria H. pylori causes ulcers,” although stronger, was not conclusively supported (p. 511). 153  By contrast, in the Umbrella case, Anne and Jim (1) have previously agreed that rain is 50% likely (2) after directly observing a cloudy sky.

138

On Presumptions, and Asymmetrical Allocations of Burden of Proof

Similarly, Koplin and Selgelid (2015) remark that PSDA “can seriously undermine the proper goals of bioethical inquiry” (p. 603) and propose making choices “supported by the best reasons” (p. 602). Most radically, Hahn & Oaksford (2007) argue that proof standards are entirely inapplicable to epistemic contexts because thresholds only arise “from the inherent link to action” (p. 48).154 Consequently, it is pointless to discuss PSDA’s legitimacy in epistemic dialogues. These scholars make a persuasive case, and, in effect, a cognitive presumption cannot entail the ‘challenger-consequence.’ Since “the truth is more likely to lie on the side of a really powerful argument” (Rescher 2006, p. 16), both standpoints are assessed by the following standard: “Accept the claim for which the available evidence is stronger” (p. 142). Accordingly, if Steve offers a stronger argument for “The Earth is disc-shaped,” Diane’s cognitive presumption “The Earth is globe-shaped” is rebutted.155 In short, flatEarther Steve cannot be committed to a higher proof standard since he cannot lose the discussion when producing stronger (or equally strong) evidence like Diane. Nevertheless, it is noteworthy that the Challenger must sometimes provide a stronger argument without being committed to the higher standard of proof. The proof standard is a means for assessing claims’ ultimate acceptability and becomes relevant in the concluding stage. However, on standard accounts, cognitive presumptions are initially acceptable propositions, epistemic “starting point[s]” (Rescher 2006, p. 50), or “basic premises” (Freeman 2005, p. 10). Potentially, the initial acceptability creates the second type of assessment criteria degree asymmetry:

4

Argument’s strength degree asymmetry (ASDA): If the User advances (cognitive) presumption p, and the Challenger advances contradictory standpoint q, then: Once argument A makes p initially acceptable, only some stronger argument B can make q epistemically acceptable. To illustrate this asymmetry, let us go back to the Direction’s case. Before the discussion, both parties acknowledged that “This street leads to…” (p) is epistemically plausible due to Jill’s testimony. Consequently, to make “This street does not lead…” (q) epistemically justified, Jones must produce a stronger argument. If his argument is equally strong as Jill’s testimony, a suspension of judgment is called for, and q remains unjustified. So, what was once sufficient to make p (initially) acceptable is insufficient to make q acceptable. Although the ultimate proof standard remains the same—to win in the 154 

Thresholds are needed only in (some) deliberation dialogues because “an action imposes a binary outcome (‘‘do’’

or ‘‘do not do’’) which has to be reconciled with possibly graded degrees of conviction” (Hahn & Oaksford 2007, p. 44). However, in epistemic dialogues, “it is entirely natural to hold intermediate degrees of conviction with regards to a claim” (p. 47). So, thresholds are inapplicable because, in epistemic contexts, degrees of conviction do not require reconciliation with a binary outcome. 155 

Objectively speaking, Steve will not be able to produce a stronger argument. But unlike Jim in the Umbrella case,

Steve is committed to the same evidence evaluation rules as the presumption’s proponent. What makes his dialectical position difficult is not a higher evaluation standard but a highly controversial, ‘unprovable’ standpoint.

139

CHAPTER 4

concluding stage, each party must have/provide a better argument than the other party— cognitive presumption includes a “pre-occupation of the ground” (Whately 1963, p. 112) and requires a more substantial argument from its opponent. Consider an analogy. Two athletes intend to break the world high jump record, and their jumps are assessed by the identical standard: to win, an athlete must achieve the best result ever recorded. Suppose that Athlete 1 goes first and succeeds: he jumps over 2.47 m, breaks the previous record (2.45 m) and becomes the all-time best. This affects the Athlete 2’s position in a straightforward way: his task immediately becomes more demanding. To achieve what her competitor has just achieved (to become the all-time best) Athlete 2 must score a better result (>2.47 m) than Athlete 1 needed to score at the start of the competition (>2.45 m). Thus, just like the order of performance in an athletic competition sometimes affects the difficulty of Athlete 2’s task, the order of defence in discussion sometimes affects the weight of the Challenger’s burden although, in some sense, the proof standard (‘The stronger argument wins’) remains the same. So, the order of defence and the burden’s allocation are not as sharply distinguished as some scholars—e.g., van Eemeren & Houtlosser (2002)—would have it. In Section 3, I argued that, sometimes, presumptions entail the same binary asymmetry (LBBA), but in distinct dialogical settings. The message of Section 4.1 is the following: different presumptions may entail distinct degree asymmetries. While cognitive presumption does not entail PSDA, practical presumption p entails PSDA in two settings: (1) when p’s rationale is not transparent, and (2) when the Challenger sufficiently defeats p in the argumentation stage. As long as PSDA and ASDA are concerned, practical and cognitive presumptions have different deontic functions.

4.2 Evidence Accessibility Degree Asymmetries Next, even if all evaluation criteria are symmetrical, the Challenger’s burden can be more demanding because (strong) arguments for the Challenger’s standpoint are less accessible than (strong) arguments for the User’s presumption. Let us illustrate the corresponding evidence accessibility degree asymmetry (EADA) on the Flat Earth example.156 How can flat-Earther Steve carry a heavier burden if Steve and Diane are committed to equal evidential requirements? First, while Diane can easily select many strong, ready-made arguments, Steve might need to construct some. Diane can justify “The Earth is globe-shaped” without being especially creative. Among many strong and accessible reasons supporting her standpoint, she can pick an adequate one easily. For instance, since Steve does not trust scientists, she can use an argument resting entirely on observation: 156 

Scholars use various examples to illustrate this type of degree asymmetry. Dare & Kingsbury (2008, p. 515)

explain that, due to better access to information, the taxpayer (instead of the tax collector) must prove the exact level of taxable income. Similarly, in cases of product liability, the burden rests with the manufacturer of a faulty product (Hahn & Oaksford 2007, p. 41).

140

On Presumptions, and Asymmetrical Allocations of Burden of Proof

(t1) Diane: Presumably(c), the Earth is globe-shaped. (t2) Steve: No, the Earth is disc-shaped. (t3) Diane: But Steve, didn’t you notice that the ship ‘sinks’ below the horizon as it sails towards it? By contrast, since, nowadays, well-known arguments for “The Earth is disc-shaped” are refuted, picking a strong argument for Steve’s standpoint is a difficult task. To offer an undefeated argument, he must conduct more in-depth research or articulate the argument himself. Also, Steve will have trouble articulating an adequate reason since Diane, by supposition, rejects the conspiracy theory supporting the flat Earth hypothesis. So, in this particular sense, his burden is more demanding: it takes more skill, creativity, and effort to defend controversial standpoints with persuasive (relevant, strong, and adequate) reasons. This explains the first kind of EADA. EADA 1: If the User advances presumption p, and the Challenger advances contradictory standpoint q, then acquiring (constructing) a persuasive argument for q is more demanding than acquiring (constructing) such argument for p.

4

Note that the asymmetry in EADA 1 is a matter of private, intrasubjective accessibility: the persuasive evidence is less accessible to the burdened party (the challenger Steve). But Steve might shoulder a heavier burden even if he had a direct epistemic access to persuasive reasons. Suppose Steve went on an expedition, and government agents stopped him just before reaching the edge of the Earth. In this distant possible world, he will have a more demanding probative job than Diane because his evidence does not represent common knowledge. Since Diane cannot be familiar with Steve’s new, ground-breaking findings, she will have many questions. When did the expedition start? Which route did it take? How did you recognise the Earth’s edge? Is there material evidence proving that government officials stopped you? Are there other witnesses, etc.? While Diane might refer to many of her arguments casually (since they are well-known and familiar to Steve), Steve must do a lot of presenting. The asymmetry in public, intersubjective accessibility forces him to “provide more evidence” (Allen 2004, p. 81) and “do more supporting” (Aijaz et al. 2013, p. 260) than Diane, and makes his dialectical burden heavier in the following sense. EADA 2: If the User advances presumption p, and the Challenger advances contradictory standpoint q, then presenting a persuasive argument for q is more demanding than presenting such argument for p. In Dare & Kingsbury’s words, there might be a “difference in what is required to discharge an equal burden” and, typically, this difference explains why proponents of “new theories will have more work to do in support of their view” (2008, p. 511). Thus, instead of committing proponents to reach unequal finish lines (like PSDA or ASDA), EADA recognises that proponents might start from unequal starting lines.

141

CHAPTER 4

Cognitive presumptions naturally entail EADA since they are typically based on common knowledge (“The Earth is globe-shaped,” “Smoking is unhealthy,” “Cereal is not a soup,” etc.), or wear justifications on their sleeves (“The cat is in a tree”). But evidence accessibility can also be associated with practical presumptions. For instance, finding the evidence for “John is guilty” in a criminal trial may be challenging if the crime happened a long time ago (Walton 2014, p. 1), or the evidence is supposed to surface in a distant future (Aijaz et al. 2013, pp. 266-267). Such cases include both PSDA and EADA 1. Next, suppose Mark justifies “Presumably(p), John will attend the party” by relying on some widely-known reason, while Alice supports “John will not attend the party” with reason that conflicts with general knowledge. (t3) Mark: John is in town, and we both know he likes to party. (t4) Alice: But John attends a sewing course on Saturday evenings! Clearly, Alice is about to face a lot of questions, regardless of whether her justification is plausible ultima facie. Are you talking about our John? Where did you get this strange information? Is the sewing class really on Saturday evening? Is it every Saturday evening? Since Alice, unlike Mark, must provide a detailed and extensive presentation of her reasons, Mark’s practical presumption “John will attend the party” entails EADA 2.

4.3 Commitment’s Strictness Degree Asymmetries Finally, the Challenger can have a heavier burden even if parties have an equally difficult job of presenting arguments, and, ultimately, their arguments are evaluated by the identical standard. That is, apart from an obligation to do more, a heavier burden might mean a stricter obligation to do the same, thereby leading to the commitment’s strictness degree asymmetry (CSDA). CSDA: If the User advances presumption p, and the Challenger advances contradictory standpoint q, then the Challenger is more strictly committed to present an argument for q than the User for p. Philosophers and argumentation scholars acknowledge that the commitment’s strictness partly determines the burden’s weight. Aijaz et al. (2013) maintain that carrying a heavier burden “is to be either under a greater obligation, or obliged to do more supporting” (p. 260), and Cargile (1997) stresses that the “strength or degree of the obligation should not be confused with the difficulty of fulfilling it” (p. 68). However, they do not develop sophisticated accounts of the obligation’s strictness. When analysing the burden’s uneven distributions, Aijaz et al. (2013) set the notion of commitment’s strictness aside, and Cargile illustrates it in passing. If Bill promises to deliver 100lbs of beans to Boyd by noon, and also promises to deliver 20lbs to Bob by noon, he has, in a sense, a lighter obligation to Bob. … But 142

On Presumptions, and Asymmetrical Allocations of Burden of Proof

if he promised Bob he would deliver the beans come Hell or high water, then he may have to wade through flood water with Bob’s beans, while Boyd has to recognize that circumstances have voided Bill’s obligation to him. (Cargile 1997, p. 68) In my interpretation, Cargile explains the commitment’s strictness in terms of prioritising one particular obligation in the circumstances of a genuine dilemma. Dilemma strictness. Commitment for Q-ing is stricter than a commitment for P-ing if Q-ing remains obligatory, and P-ing becomes cancellable once the party must choose between P-ing and Q-ing. Let us apply the dilemma strictness to dialogical settings involving CSDA. The result is the following: CSDA 1: The Challenger is more strictly committed to present an argument (for q) than the User (for presumption p) because supporting q remains obligatory, and supporting p becomes cancellable once we must choose between supporting p and supporting q. Both practical and cognitive presumptions may entail CSDA 1. Suppose time is so limited in the Party example that only one party can present the final argument. Also, suppose Mark’s arguments for “Presumably(p), John will attend the party” have become quite repetitive, and we cannot reasonably expect Mark has anything new to offer. In such circumstances, cancelling Mark’s obligation and giving Alice the last word appears dialectically desirable.

4

Similarly, suppose that in the Flat-Earth example we must choose between hearing another well-known argument for “Presumably(c), the Earth is globe-shaped” and acquiring brand-new evidence for “The Earth is disc-shaped” (e.g., evidence indicating that government officials stopped Steve at the edge of the Earth). Since Diane’s reasons belong to public knowledge, cancelling her burden is not very costly from an epistemic viewpoint. That is, a reasonable judge who wishes to assess an epistemic status of Diane’s standpoint will likely take her final (unexpressed) argument into consideration, anyway. By contrast, cancelling Steve’s burden is more costly since neither Diane nor some external judge will be able to reconstruct an unexpressed and publicly inaccessible argument. Since Steve’s argument might contribute more to the epistemic assessment than Diane’s already-known argument, he is more strictly committed to argue, and her burden is cancellable. Portraying the commitment’s strictness in previous fashion limits the analysis of asymmetrical allocations to contexts where parties face a practical dilemma. But parties are not typically forced to defend one standpoint at the expense of not defending another. The previously described dilemma is highly unusual, so the following becomes crucial: When both burdens (in principle) can be fulfilled, in the standard dialogical settings, can the Challenger’s commitment to support q be stricter than the User’s commitment to support p? Consider the next formulation:

143

CHAPTER 4

Rationality-undermining strictness. Commitment for Q-ing is stricter than a commitment for P-ing if the failure in Q-ing undermines the rationality (of a process, or procedure) more seriously than the failure in P-ing. Merging the rationality-undermining strictness with CSDA leads to the following result: CSDA 2: The Challenger is more strictly committed (i.e., obliged) to present an argument for q, while the User is less strictly committed (i.e., recommended) to submit an argument for presumption p because the failure in supporting q undermines the discussion’s rationality more seriously than the failure in supporting p. As showed in Section 4.1, Mark should support “Presumably(p), John will attend the party” when his rationale is not transparent. But even if his rationale is not entirely transparent to Alice—e.g., in the sense that she has explicitly conceded Mark’s argument at some previous time—it can still be easily accessible. Suppose Mark believes John might attend the party because “Generally, John likes to party”. Most likely, Alice will consider this justification spontaneously for it concerns the character of the person she knows, and they would not be discussing John’s attendance if John did not like to party.157 In such circumstances, the implicit nature of Mark’s argument does not seriously undermine the deliberation’s rationality. By contrast, making the right decision may be seriously threatened if Alice failed to present evidence that John will attend Saturday’s sewing course. So, according to CSDA 2, she is more strictly committed to arguing than Mark. Under comparable evidential conditions, this result may apply to cognitive presumptions as well. We can further explain the commitment’s strictness by identifying the kind of dialectical rule underlying the party’s burden of proof. For instance, Alice’s commitment to defend “We should attend the party” is based on the elementary rule of critical discussion: “Discussants who advance a standpoint may not refuse to defend this standpoint when requested to do so” (van Eemeren & Grootendorst 2004, p. 191). Since the rules of critical discussion “determine whether or not a fallacy has been committed” (van Laar & Krabbe 2013, p. 204), refusing to defend a standpoint is fallacious and, therefore, most seriously undermines the deliberation’s rationality. Consequently, Alice has a very strict probative commitment. By contrast, Mark’s commitment to support “Presumably(p), John will attend the party” is governed by the optimality rule that “distinguishes between non-fallacious moves of higher and of lower quality” (van Laar 2014, p 407). Since the parties should “Choose the clearest formulation” (Krabbe & van Laar 2011, p. 225) and the explicit formulation is clearer than the implicit one, Mark is recommended to present an explicit argument. However, by failing to do so, he does not commit a fallacy. Sometimes, keeping an easily accessible argument to oneself is a move of a lower quality, but hardly “obstructs or hinders the resolution of a difference of opinion on the merits” (van Eemeren et al. 2014, p. 157  As Goldman (2003, p. 60) suggests, while explaining the epistemic approach to enthymemes: “In assessing a conclusion’s plausibility, the hearer of an argument does not rely exclusively on the speaker’s stated premises. A thoughtful hearer will marshal all relevant evidence at her disposal, at least all evidence that readily comes to mind.”

144

On Presumptions, and Asymmetrical Allocations of Burden of Proof

545). Accordingly, Mark’s probative commitment is not very strict: while rules of critical discussion make Alice obliged to argue, optimality rules only recommend Mark to do so. Does this imply that the commitment strictness is always correlated to evidence accessibility? The intersubjective accessibility of the User’s reasons appears to generate two asymmetries: the User has an easier job presenting her argument (EADA 2), and the User is only recommended to provide an argument (CSDA 2). However, the strictness of probative commitment also depends on the standpoint’s illocutionary force, i.e., a particular speech-act the speaker is putting forward (see Kauffeld 1998, 2003, 2005, 2019). For instance, the speech-act of “proposing” (2005, p. 5) entails the stricter commitment to present reasons than the speech-act of “suggesting” (1998, p. 247) or “advising” (2005, p. 5). In other words, the speaker undermines the communication’s rationality more seriously if she refuses to support her proposal, than if she refuses to support her suggestion. This shows that CSDA 2 is not necessarily linked to justification’s transparency (accessibility), but is accompanied by its own set of conditions. Although it can be applicable to all presumptions, as long as it is correlated to EADA 2, CSDA 2 is typically connected to cognitive presumptions.

4

5 Conclusion This chapter analysed the unique deontic function of presumption by focusing on three interrelated objectives: 1. Developing a comprehensive account of the burden’s ‘asymmetrical allocation’ (conceptual objective); 2. Exploring (a) in which senses presumptions allocate the burdens asymmetrically, (b) in which dialogical settings, and (c) are these senses and settings typical to each type of presumption (dialectical objective); 3. Comparing the deontic functions of practical and cognitive presumptions (typological objective). As far as the conceptual objective is concerned, this chapter identified, labelled, systematically presented, and located in the broader semantic network asymmetries that have already been explored in the literature (LBBA, HBBA, PSDA, EADA 1). Crucially, it investigated some asymmetries which have lacked a detailed analysis so far (ASDA, EADA 2, CSDA). The scheme below (Figure 1) offers a systematic overview of different burden asymmetries.

145

CHAPTER 4

As far as dialectical and typological objectives are concerned, the chapter offers four results. 1. Sometimes, the two types of presumptions entail the same dialectical asymmetry (LBBA, HBBA), but in different dialogical settings. For instance, in the setting where the User advances presumption p at t1, and the Challenger sufficiently defeats p at tn: • At tn+1, the User is obliged to support practical presumption p (“It will rain”); • At tn+1, the User is not obliged to support cognitive presumption p (“This street leads to the city center”). 2. Even if presumptions entail similar asymmetries (e.g., HBBA), their challengers might carry distinct burdens of proof. For instance, while the Challenger of practical presumption carries the burden of persuasion, the Challenger of cognitive presumption carries the burden of successful defence in the concluding stage. 3. Some burden asymmetries are associated only with a particular type of presumption. For instance, only the Challenger of practical presumptions is required to meet the higher proof standard. So, practical presumption does entail PSDA, and cognitive presumption does not. 4. Even when presumptions entail the same asymmetry in similar settings, these asymmetries and settings might be more typical for one type of presumption. For instance, the conditions where evidence is easily accessible are more typical for cognitive presumptions. Propositions like “The Earth is globe-shaped,” “The cat is in a tree,” or “Smoking is unhealthy” usually wear justifications on their sleeves. In contrast, practical presumptions are often used for reducing the risk in deliberation when there is “the absence of evidence on [the] issue” (Ullmann-Margalit 1983, p. 146). Accordingly, cognitive presumptions typically entail EADA whereas practical presumptions only do so occasionally. This chapter sends two general messages to presumption theorists. First, whether presumption p allocates dialectical burden asymmetrically often depends on p’s inessential features, such as transparency, or accessibility of p’s foundations. Since traditional scholars view the burden’s asymmetrical allocation as presumption’s defining function, this result is theoretically puzzling. Second, practical and cognitive presumptions are relevantly different even at the most fundamental level of deontic function: contrary to Rescher’s suggestion, cognitive presumptions are not “the epistemic analogue[s] of ‘innocent until proven guilty’” (2006, p. 23), and the two classes of presumptions merit separate treatment.

146

On Presumptions, and Asymmetrical Allocations of Burden of Proof

Acknowledgments: The earlier version of this paper was presented at the ARGA colloquium (University of Amsterdam, Department of Speech Communication, Argumentation Theory and Rhetoric), on March 26 2021. I thank the organizers of this colloquium, as well as the participants for their comments and suggestions. I am especially indebted to my supervisors dr. Jan Albert van Laar and dr. Barteld Kooi, and Karolina Kudlek.

4

Figure 1 Asymmetrical allocations of dialectical burden: A systematic overview

147

5

CHAPTER 5 On the Strength of Presumptions

CHAPTER 5

Your strength is just an accident arising from the weakness of others.

―Conrad, Heart of Darkness

150

On the Strength of Presumptions

On the Strength of Presumptions Abstract. Traditionally, a presumption is a dialogically privileged, yet defeasible proposition that allocates the burden of proof to a party who challenges it. This chapter investigates the strength of presumptions. First, it explains how ‘strength’ contributes to defining the concept of presumption. Second, it provides an overview of (contextual, justificatory, and deontic) factors determining a presumption’s strength. Finally, it analyses the predominant view that defines strength in terms of the Challenger’s burden of proof: the stronger (weaker) the presumption p, the more (less) difficult it is to prove non-p. I argue that the latter proposal applies only to practical presumptions, and that strength is conceived differently for cognitive presumptions. Keywords: burden of proof, cognitive presumption, plausibility, practical presumption, rebuttal, risk, strength, synchronic-diachronic, values.

1 Introduction Traditionally, philosophers and argumentation scholars understand presumptions as propositions introduced with the modal qualifier ‘presumably’ (Ullmann-Margalit 1983; Hansen 2003; Rescher 2006; Walton 2014; Godden 2017, 2019). “Presumably, p” indicates an agent’s provisional commitment to take p as true given some presumptive rule and a basic fact.

5

Suppose Emma deliberates whether to wear a face mask during the COVID-19 pandemic. Suppose that, after assessing all relevant evidence, it remains uncertain for Emma whether wearing a mask protects other people from becoming infected (basic fact). In such conditions, Emma should exercise caution and abide by a presumptive rule: “If it is uncertain whether wearing a mask protects other people from contracting COVID-19, then, all else being equal, a deliberating agent should act as if it does.” Emma thus becomes committed to a practical presumption: she should proceed on “Presumably(p), wearing a face mask protects other people from contracting COVID-19” until sufficient evidence determines that a mask is useless. Accordingly, practical presumptions promote non-epistemic goals (such as precaution or social safety) in the context of evidential uncertainty and deliberative pressure (Ullmann-Margalit 1983; Godden 2017, 2019). Next, suppose Henry sees a man who looks exactly like Tom, walking a dog in the street (basic fact). Due to the reliability of perception, he should abide by the presumptive rule: “If visual perception indicates that Tom is walking a dog, then, all else being equal, a cognitive agent should accept that ‘Tom is walking a dog’.” Henry thus becomes committed to a cognitive presumption: he should proceed on “Presumably(c), Tom is walking a dog” until the statement is proven false or unjustified. On standard accounts, cognitive presumptions promote epistemic goals, such as attaining truth or justified belief.

151

CHAPTER 5

They are warranted by common epistemic sources (perception, memory, testimony, etc.) and represent the most plausible propositions (Rescher 2006; Freeman 2005). Supposedly, presumptions asymmetrically allocate the burden of proof. After rejecting “Presumably, p,” the Challenger must substantiate her rejection, while, typically, the presumption’s User (proponent) is not immediately obliged to argue in favour of “Presumably, p.” For instance, if Tom rejects “Presumably(p), wearing a face mask protects other people from contracting COVID-19,” he must provide reasons. If his reasons are weak, Emma’s presumption is acceptable, and she is not obliged to defend it. Similarly, if Mia does not provide sufficient argument against “Presumably(c), Tom is walking a dog,” Henry’s presumption does not require further support. Hence, a presumption stands unless the Challenger defeats it with sufficient argument (Whately 1963, p. 112; UllmannMargalit 1983, p. 150; Pinto 2001, p. 4; Freeman 2005, p. ix; Rescher 2006, p. 15; Godden & Walton 2007, p. 334). According to Godden & Walton (2007), a fully-fledged theory of presumption must elaborate the force of presumptions. In this chapter, I explore the ‘strength’ of presumptions by addressing three sub-questions: 1. The most general sub-question: What is the role of strength in a theory of presumption? Section 2 identifies two roles. First, ‘strength’ may frame the concept of presumption (conceptual research). Second, one may associate ‘strength’ with factors that make a presumption p relatively stronger or weaker (strength research). 2. The less general sub-question: What specific factors affect the strength of presumption? Section 3 provides a systematic overview of the three categories of factors: contextual, justificatory, and deontic factors (such as the weight of the Challenger’s burden of proof). 3. The most specific sub-question: Is a presumption’s strength always correlated with the Challenger’s burden of proof? When presumption p becomes stronger (weaker), does the Challenger’s burden to defeat “Presumably, p” become heavier (lighter)? Do practical and cognitive presumptions differ in this respect? Section 4 tackles these questions. I conclude that practical and cognitive presumptions require different theories of strength. While the force of a practical presumption is correlated with the weight of rebuttal, cognitive presumption p may become weaker despite the Challenger’s burden not becoming lighter, i.e., despite non-p not becoming easier to prove. This provides one reason for separate treatment of practical and cognitive presumptions in argumentation theory (for other reasons, see Bodlović 2020b, 2021).

152

On the Strength of Presumptions

2 Strength in Presumption Theory In contemporary scholarship on presumption, exploring strength involves two lines of research. I call these ‘conceptual research’ and ‘strength research.’ Conceptual research aims to define the concept of ‘presumption.’ Theorists focus on the following question: How does ‘strength’ (or ‘force’) explain what a presumption is? By contrast, strength research investigates conditions affecting the strength of presumptions. Here, theorists focus on the questions: What makes presumption p stronger (weaker) than presumption r, and what makes presumption p stronger (weaker) over time? This chapter contributes to strength research. In Section 2.1, however, I illustrate some positions within the conceptual project. Two meanings of ‘strength’ are particularly relevant for identifying presumption p: the strength of p’s epistemic justification and p’s pragmatic (illocutionary) force.

2.1 Conceptual Research First, a presumption is defined by the strength of epistemic justification. On standard accounts, practical presumption p is “triggered in the absence of certain reasons for belief” (Ullmann-Margalit 1983, p. 152), or the “absence of sufficient or determinative evidence that p is the case” (Godden 2017, p. 508). Similarly, Walton maintains that an argument supporting a presumption is “not sufficiently strong, based only on the evidence supporting the two premises to shift a burden of production to the respondent’s side” (2014, p. 110). Moreover, when it concerns cognitive presumptions, Rescher (2006) stresses that they are not “absolutely certain or totally self-evidencing theses” (p. 20) that come with “categorical guarantees” (p. 31). So, if “A face mask protects other people from contracting COVID-19” were a justified belief, it wouldn’t be Emma’s (typical) practical presumption, and if “Tom is walking a dog” were certain, it wouldn’t be Henry’s cognitive presumption. Thus, to identify p as a presumption, the strength of p’s epistemic justification seems essential.

5

Second, presumption is occasionally portrayed as a speech act with distinctive illocutionary force. Speech-act theorists emphasise that, in everyday communication, speakers use language for various purposes (see Austin 1962). Consequently, the same utterance can have distinct pragmatic meanings. For instance, by saying “This concert is boring” the speaker may explain how things are, but also complain about the band’s performance or (indirectly) propose a line of action (“We should leave this place”). Among other aspects, the utterance’s illocutionary force depends on the speaker’s communicative aim, as well as a conversational effect the proposition is supposed to achieve (Searle & Vanderveken 1985). What is a practical presumption’s illocutionary force? What does Emma do by saying “Presumably(p), wearing a face mask protects other people from contracting COVID-19?” According to Godden, “presumings that p do not involve commitments to

153

CHAPTER 5

p’s truth or assertions—even qualified or tentative assertions—that p” (2017, p. 508). Practical presumption facilitates action. By presuming, Emma only communicates that— due to uncertainty, pressure, and precaution—it is reasonable to proceed as if a mask provides protection. Cognitive presumptions have different illocutionary force. By saying “Presumably(c), Tom is walking a dog,” Henry expresses his belief, thereby committing to the truth of the presumed claim. However, despite this difference, practical and cognitive presumptions might have an identical or closely similar illocutionary force due to allocating the burden of proof on the presumption’s Challenger: I also assume, following Walton (1993) and Corredor (2017), that to presume something is to make a conversational move understood as an illocutionary act, the force of which is to be characterised by reference to the way it modifies the distribution of the burden of proof among the participants in a dialogue. (Witek 2019) On Walton’s (1993) view, speech acts are characterised by how they distribute the burden of proof. If the Challenger rejects a ‘presumption’ at t2, she takes up the burden of proof at t3, while the User is not immediately obliged to argue in favour of her presumption. But if the challenger rejects an ‘assertion’ at t2, the proponent must defend it at t3. Finally, if the challenger rejects an ‘assumption’ at t2, she is not obliged to justify her rejection at t3. Thus, the speech act of presuming is dialectically stronger than the speech act of assuming: while the challenger can reject the assumption without offering a reason in return, she can successfully reject the presumption only by offering a persuasive argument. As Godden & Walton put it, unlike presuming, “assuming bears no cost, but has no force” (2007, pp. 326–327). As a result, the concept of illocutionary force serves a double purpose: it describes essential features of various speech acts and enables us to compare their dialectical strengths. This concludes my illustration of the usual functions of ‘force’ or ‘strength’ within conceptual research. In the following sections, I systematically explore factors that influence the strength of presumptions.

2.2 Strength Research As Whately observes, “[a] presumption evidently admits of various degrees of strength, from the very faintest, up to a complete and confident acquiescence” (2019, p. 49). Analysing this variety lies at the heart of strength research. However, on closer inspection, strength research can focus on two phenomena and involve two sub-approaches. I call them the ‘synchronic’ and the ‘diachronic’ approach. On the synchronic approach, the theorist explores conditions that make presumption p stronger or weaker at some fixed point in time. She can analyse p at some time t while explaining why p is stronger in context C1 than in context C2. Furthermore, she can analyse presumptions p and r at t while explaining why p is stronger than r in context C.

154

On the Strength of Presumptions

By contrast, on the diachronic approach, the theorist investigates conditions that make presumption p stronger or weaker over time. For instance, she can measure p’s strength at t1 and t2 and identify factors that render p stronger at t2 than at t1. In Section 3, I explain factors that affect the strength of a presumption, relevant for both the synchronic and the diachronic approaches. Although these factors are already acknowledged in the literature, they are discussed under different headings (e.g., in the studies of fallacies, burdens of proof, etc.) and are not systematically reviewed in one place. In this chapter, I bring them all together and explicate their effect on the strength of presumptions.

3 What Determines the Strength of a Presumption? Presumptions are stronger or weaker, heavier or lighter, depending on contextual, justificatory, and deontic factors. The three classes of factors are interconnected, and the boundaries between them blurry. However, distinguishing the three classes contributes to presentational clarity and a clear-headed analysis of the strength of presumptions.

3.1 Contextual Factors What is the effect of contextual factors on the strength of presumptions? More precisely, how do the features of dialogue make presumption p stronger or weaker? And what features are we talking about?

5

The leading dialecticians in argumentation theory argue that an argument’s strength, as well as its fallaciousness is determined by the type of dialogue (Walton & Krabbe 1995). The rhetoricians say that an argument’s strength—in the sense of its rhetorical effectiveness—varies between different audiences (Perelman & Olbrechts-Tyteca 1969, pp. 460-471). Consequently, presumption p can receive stronger or weaker support depending on dialogue type or rhetorical context, or, as Godden & Walton (2007) call it, the “argumentative situation” (p. 315), or “circumstantial conditions” (p. 338).

3.1.1 The Dialogue’s Collective Goal Several features characterise the type of dialogue, the central one being the dialogue’s “main goal” (Walton & Krabbe 1995, p. 66), “overall goal,” or “collective goal” (Walton 1998, p. 30). Epistemic dialogues, which aim to produce knowledge, differ from practical dialogues, which serve to choose the right action. The argument’s quality, then, depends on the nature of the collective goal (Walton 1998, p. 30). For instance, assuming that “Presumably, p” contributes to choosing the right action, it will be strong if introduced in practical dialogue (which I call ‘deliberation’), but possibly weak if presented in the context of a purely epistemic dialogue (which I call ‘inquiry’).158 158 

My use of the terms ‘deliberation’ and ‘inquiry’ is different from how they are typically used (see, e.g., Walton &

Krabbe 1995; Walton 1998, Atkinson et al. 2013).

155

CHAPTER 5

To illustrate, let’s analyse the practical presumption of innocence—the paradigmatic presumption in criminal law—from the synchronic perspective. The aim of proceeding as if the accused is innocent (until proven guilty) is reducing the risk of ‘false positives’ (Bermejo-Luque 2016, p. 15): since the prosecutor carries the burden of persuasion, she must prove guilt beyond a reasonable doubt, which significantly lowers the risk of convicting an innocent person. The presumption of innocence is grounded in the hierarchy of values that treats convicting an innocent person as more harmful than letting a guilty person go free (Dare & Kingsbury 2008, p. 507). Since mistakes are inevitable, given evidential uncertainty, parties should avoid costlier errors. Thus, the presumption of innocence promotes a higher non-epistemic value, reduces the chance of making a costlier error, and thereby gains in strength in legal deliberation C1. Nevertheless, the presumption of innocence becomes weaker—if it were even applicable—in the context of epistemic inquiry C2. Namely, a person would hardly be accused to begin with if some evidence did not indicate her guilt. Since the accusation is not a solid epistemic indicator of innocence, the presumption of innocence does not promote the attainment of knowledge (Ullmann-Margalit 1983, p. 158; Rescher 2006, p. 9), and for this reason becomes weak or even inapplicable in C2. The synchronic approach also offers an alternative way of analysing the contextdependency of presumption strength: at some fixed time t, presumption p can be stronger than presumption r due to the dialogue’s collective goal. The collective goal’s explanatory function remains the same: just as p is stronger in context C1 than in context C2 if p contributes more to the collective goal of C1 (than to the goal of C2), p is stronger than r in C if p contributes more (than r) to C’s collective goal. Suppose Henry and Mary want to know whether Tom is in town. A friend says to Mary: “Tom left town. He talked about taking a family trip over the weekend.” Given the rule “Accept at face value the declarations of other people” (Rescher 2006, p. 31), Mary infers “Presumably(c), Tom is not in town.” But imagine that Henry sees a person (walking a dog) that looks exactly like Tom. Since “To believe the evidence of your own senses” (Rescher 2006, p. 31) represents a basic rule of cognitive presumption, Henry infers “Tom is walking a dog” and concludes “Presumably(c), Tom is in town.” Intuitively, in epistemic contexts, where resolving an issue does not require expertise, a presumption supported by first-hand evidence (direct observation) will get us closer to the truth—i.e., the dialogue’s collective goal—than the one supported by second-hand evidence (testimony). Thus, Henry’s cognitive presumption is stronger than Mary’s: in inquiry C, at time t, presumption p is stronger than presumption r.159

159  In this example, C represents an epistemic inquiry as a dialogue type, not as a particular instance of a dialogue type. In other words, Henry and Mary are not participating in the same dialogical exchange—they are not talking to each other. If they were, we wouldn’t be analysing two presumptions with distinct degrees of strength. Instead, we would be analysing why Henry’s cognitive presumption—as the most plausible truth-candidate (Rescher 2006)—defeats Mary’s standpoint.

156

On the Strength of Presumptions

3.1.2 The Dialogue’s Cooperativeness Dialogues also differ in terms of cooperativeness (see Walton 1998). Accordingly, presumption p may be strong or weak depending on whether dialogue C is predominantly cooperative or adversarial. Consider a dialogue fragment initially associated with the Bronston v. U.S. case (Macagno 2012, p. 287). During the course of the dialogue, an issue arises concerning Dave’s bank accounts: Peter: Have you ever had any bank accounts in Swiss banks? Dave: The company had an account there for about six months. What is the meaning of Dave’s answer? Should Peter presume that Dave has never had any accounts in Swiss banks? The answer depends on the degree of cooperation between the parties. Suppose parties participate in a cooperative dialogue C1 where offering conflicting arguments is a means to reach a mutual goal (such as accepting the most plausible standpoint) and there is no ‘bad blood’ between the parties. In that case, Peter should interpret Dave’s answer charitably. He may presume that David follows Grice’s Cooperative Principle and aims to contribute to the discussion in a truthful, informative, relevant, and clear manner (Grice 1975, pp. 45-6). Given that participants appreciate relevance, “Dave has never had any accounts in Swiss banks” seems a natural interpretation of Dave’s answer in C1.

5

However, the previous exchange originally belongs to an adversarial legal context C2. Since the prosecutor and the defendant typically cooperate to a minimal degree, the prosecutor (Peter) should not apply the rules of charitable interpretation easily in C2. Instead, he should presume that the defendant (Dave) will deliberately offer unclear, irrelevant, or uninformative answers when facing difficult questions (Goodwin 2000, p. 265; Macagno 2012, p. 257). In other words, instead of appealing to the Cooperative Principle, the prosecutor should follow the “noncooperative Principle of Zeal” (Goodwin 2000, p. 263), suspend judgment on the meaning of the defendant’s message, and demand further clarification. Consequently, the cognitive presumption “Dave has never had any accounts in Swiss banks” is weaker (if it is even applicable) in C2 than in C1. Thus, presumption p will become weaker or even lose its presumptive status in an adversarial dialogue whenever cooperation is crucial for presuming p. Let’s illustrate this in a diachronic fashion as well. Suppose that a cooperative persuasion dialogue C1 shifts into an eristic dialogue C2. Given this ‘dialectical shift’ (Walton & Krabbe 1995), Peter might rightfully reassess the strength of “Presumably(c), Dave has never had any accounts in Swiss banks” in C2. Plausibly, the presumption will be weaker in C2, at t2, than it was in C1, at t1.

157

CHAPTER 5

3.1.3 Risks, Stakes, and Background Information The local features of dialogue also affect a presumption’s strength. For instance, deliberation dialogues C1 and C2, with identical collective goals (‘choosing the best action’) and similar degrees of cooperation, might differ in terms of practical stakes and risk assessment. Also, epistemic inquiries C1 and C2, with identical collective goals (‘accepting an epistemically justified proposition’) and similar degrees of cooperation, might differ in terms of background information. Consequently, if dialogues differ in terms of stakes or information, then presumption p may be stronger in C1 than in C2. As explained in Section 3.1.1, the presumption of innocence reduces the risk of convicting innocent people. In criminal law deliberation C1, the stakes are high because imprisoning an innocent man (John) violates his legal rights and ruins his life. But innocence can also be presumed in some non-legal deliberation C2. Suppose Peter blames a co-worker John for stealing his favourite pencil and decides to expose John to their colleagues if the evidence indicates guilt. However, after the investigation, John’s responsibility remains uncertain, so Peter presumes innocence: given insufficient evidence to the contrary, he acts as if John did not steal his pencil. In conditions of uncertainty, leaving such a minor incident unpunished is less costly than embarrassing an innocent person. One man’s reputation is more important than another man’s pencil. Imprisoning an innocent man is more costly than criticising him, so, due to particularly high stakes, “Presumably(p), John is innocent” is stronger in C1 than in C2. As a result, in C1, the Challenger must present stronger arguments to defeat the presumption: while the prosecutor must prove guilt ‘beyond a reasonable doubt’ in C1, Peter can defeat “Presumably(p), John is innocent” by offering much weaker evidence in C2. For instance, imagine Peter’s co-worker saying: “Earlier today, I saw John carrying a similar pencil.” Since stakes are lower, such testimony may suffice to stop acting like John was innocent and to criticise him. How can background information affect the strength of a presumption? Consider a modification of Kauffeld’s (2009) example where Bruce promises to return a borrowed item. By doing so, Bruce becomes committed to behave in a particular way, and we can expect him to honor this commitment “rather than risk criticism, painful regret, reprobation, loss of esteem or even punishment for failing to do so” (Kauffeld 2003, p. 140). In the epistemic context C1, his promise allows us to come to a reasonable expectation “Presumably(c), Bruce will return the borrowed item.”160 But consider context C2 where Bruce also leaves a deposit. Since ignoring his promise is more costly in C2, Bruce is more likely to return the item in C2, and our cognitive presumption “Bruce will return the borrowed item” becomes stronger in C2 than in C1. As Kauffeld explains: “[P]resumptive inference is weakened or strengthened by variation in the circumstances entering an agent’s calculation of the risk of resentment he would run were he to fail to act in accord with the commitments warranting the presumption” (2009, p. 8). 160  Kauffeld (2003) defines presumption as an expectation based on social commitment. Thus, one can presume p if one is entitled to expect p based on someone’s responsibility or obligation to make p true.

158

On the Strength of Presumptions

It is controversial whether a theorist should explain such variations in strength in terms of context or justification. Stakes and costs can be treated as contextual features but also as premises of practical reasoning that directly justify “Presumably(p), John is innocent” (Bodlović 2021, pp. 294-295). Similarly, “Bruce left a deposit” can be treated as contextual information increasing the credibility of Bruce’s justification (“I will return the item because I promised to do so”), but also as an equally significant, additional component of justification (“I will return the item because I promised to do so, but also because I’ve got money to lose”). My own assessment is that interpretations are equally plausible as long as the theorist considers all relevant information. Clearly, an argument’s content and structure can remain unchanged even if its relational properties—such as strength—vary from one context to another.161 But sometimes, a change in context modifies an argument’s content and structure, thereby directly affecting its strength. In such cases, the distinction between contextual and justificatory factors is mostly analytic.

3.1.4 The Intended Audience Finally, dialogical contexts are characterised by the audience’s beliefs, values, and competencies. Presenting an argument for “The Earth is globe-shaped” to a primary school student, a natural scientist, and a flat-Earther entails presenting the same argument in three markedly different contexts.

5

The intended audience determines the strength of Emma’s practical presumption “Presumably(p), wearing a face mask protects other people from contracting COVID-19.” If her audience consists of virologists and epidemiologists (context C1), Emma’s presumption is very strong.162 If her audience consists of the general public (C2), the presumption is moderately strong: even if it is uncertain whether masks are helpful, many people will wear them due to their potential to reduce viral transmissions. But if Emma’s audience consists of COVID-19 deniers (C3), “Wearing a face mask protects…” is rhetorically weak and may not even count as a presumption in C3. As Whately claims, some paradigmatic presumptions (such as the one “against a paradox”) favour the prevailing opinion, i.e., beliefs and values shared by the audience (2019, p. 47). Thus, even if reasons strongly justify p, p’s presumptive status is weak or nonexistent if the intended audience rejects these reasons. Rhetorically strong presumptions are backed by reasons that cohere with the audience’s commitments, enabling effective persuasion (Macagno & Damele 2013, p. 375). In short, the strength of a presumption depends on the 161 

For instance, the same set of premises and conclusions may become stronger or weaker when persuasion dialogue

C1 shifts to eristic dialogue C2. Such an argument undergoes a Cambridge change. 162  In fact, since virologists and epidemiologists agree that wearing a face mask reduces viral transmissions—i.e., since they do not treat Emma’s presumption as epistemically uncertain—“A face mask protects other people from contracting COVID-19” may have the status of cognitive, rather than practical presumption in C1.

159

CHAPTER 5

justification’s objective sufficiency, but also on the justification’s adequacy (the presumption’s justification is acceptable to the Challenger) and legitimacy (the presumption’s justification is acceptable to the Challenger’s peers) (Bodlović 2020b, pp. 267-268).

3.2 Justificatory Factors However, the strength of a presumption primarily depends on its justification, “foundation” (Godden & Walton 2007, p. 315) or “grounding” (Godden 2011, p. 3). As UllmannMargalit puts it: “[T]he factors that determine the differential strength of presumptions … have to do with the relative strength of the considerations in which the justification of each presumption is grounded” (1983, p. 154). Accordingly, the strength of a practical presumption is (mostly) grounded in “nonepistemic values or goals” (Godden 2017, p. 508), while “the strength of [typical cognitive] presumption is clearly in line with the matter of evidential weight” (Rescher 2006, p. 66). Several considerations influence a presumption’s justification. This section develops a systematic account of such factors while focusing specifically on factors concerning the presumptive rule. These factors are acknowledged in the literature, but here I specify their connection with the strength of presumptions.

3.2.1 The Basic Fact According to judicially inspired accounts, a presumption follows from a ‘basic fact’ and a ‘presumptive rule’ (Ullmann-Margalit 1983; Hansen 2003; Rescher 2006; Walton 2014; Godden 2017). The basic fact triggers presumptive reasoning by instantiating the rule’s antecedent. In the COVID-19 case, the basic fact “It is uncertain whether wearing a face mask protects other people from contracting COVID-19” supports “Presumably(p), wearing a face mask protects other people from contracting COVID-19.” In the so-called Umbrella case, the fact “Rain is uncertain” supports “Presumably(p), it will rain” and instructs Anne to bring an umbrella (Bodlović 2021). The presumption’s strength depends on the basic fact’s further justification. Since basic facts express uncertainty about presumptive practical reasoning, they are assessed by means of various considerations regarding uncertainty. Räikkä (1997) distinguishes between several kinds of uncertainty. The ‘consensual method’ treats p as an open question if “parties have [not] achieved agreement on the solution of the question” (p. 470). By contrast, the ‘rational method’ treats p as uncertain if scientific research fails to make p “objectively or intersubjectively justified” (p. 470). In my view, previous methods reveal distinct justifications of basic facts, thereby producing facts (and presumptions) of unequal strength. Plausibly, “It is uncertain whether wearing a mask protects…” is stronger if uncertainty is justified by science than if explained only by dialectical disagreement. For instance, scientific evidence showing that masks might be

160

On the Strength of Presumptions

useless would support our basic fact more strongly than a disagreement in the public sphere. The strength of a basic fact depends on the reliability of an epistemic source. In the Umbrella case, “Rain is uncertain” is stronger if justified by a trustworthy weather forecast (“The chance of rain later today is 50%”) than if justified by direct observation (“I see a cloudy sky, so it might rain later today”) or mere possibility (“There isn’t a cloud in the sky, but the weather can always change, so…”). In presumptive cognitive reasoning, a basic fact’s strength involves similar considerations.

3.2.2 The Presumptive Rule The presumptive rule has a conditional form “If basic fact q, then ‘Presumably, p’.” But what aspects of a presumptive rule make “Presumably, p” weaker or stronger? I analyse four aspects: (1) the rule’s general structure, (2) its formal status, (3) further justification, and (4) the content of the rule’s non-defeater clause. The first aspect concerns the rule’s general structure. Legal scholarship distinguishes between rebuttable and irrebuttable presumptions (Gama 2017, p. 561; Macagno & Walton 2012, p. 280). Rebuttable presumptions are justified by rules containing the ‘non-defeater clause.’163 This clause, which could be expressed with the ‘all else being equal’ phrase, indicates defeasibility: “Presumably, p” follows from a basic fact q tentatively, i.e., absent defeaters. Defeasible presumptive rule: If q, then—all else being equal—“Presumably, p.”

5

(If q, an agent should act on p until or unless…). Paradigmatic practical presumptions are rebuttable. For instance, the accused is presumed innocent until proven guilty. We should wear face masks to fight COVID-19 unless wearing masks is proven useless. Sometimes, however, presumptions cannot be defeated: if basic fact q is established, an agent should always act on p. Accordingly, the rule that justifies an irrebuttable presumption should not contain any reference to the possibility of defeaters. Indefeasible presumptive rule: If q, then (necessarily) “Presumably, p” (If q, an agent should always act on p). For instance, according to a doli incapax presumption, if a child is under a particular age (seven or ten), the court must always proceed as if she cannot have criminal intent and, therefore, is incapable of committing a crime (Ullmann-Margalit & Margalit 1982, p. 442; Macagno & Walton 2012, p. 280). Irrebuttable presumptions can also be utilised in non-legal deliberations: when changing lanes, the driver should always shoulder-check, i.e., act as if something might be in her blind spot (Godden 2017, p. 507). While scholars acknowledge the possibility of irrebuttable practical presumptions (see, also, Walton 2014, p. 114), Rescher claims that “cognitive presumptions are defeasible and hold good only ‘until further notice’” (2006, p. 35). 163 

What I call the ‘non-defeater clause’ Ullmann-Margalit calls a ‘rebuttal clause’ (1983, p. 149), and Rescher a

‘default proviso’ (2006, p. 33).

161

CHAPTER 5

This structural difference indicates distinct logical standards. Rebuttable presumptions are associated with non-monotonic reasoning. But since new evidence cannot nullify irrebuttable presumptions (assuming that the basic fact remains true), they are evaluated by the norms of monotonic reasoning. Thus, it is not surprising that irrebuttable presumptions have been labelled ‘deductive’ (Walton 2014, p. 114), ‘conclusive’ (Ullmann-Margalit & Margalit 1982, p. 442; Gama 2017, p. 561; Petroski 2008, p. 391), ‘necessary’ (Gama 2017, p. 560; Walton 2014, p. 114), and ‘indefeasible’ (Rescher 2006, p. 5). Accordingly, the type of logical connection affects the strength of a presumption straightforwardly: if presumption p necessarily follows, and presumption r tentatively follows from a basic fact, then p is stronger than r. Usually, scholars illustrate this difference in synchronic terms: at t, in a legal context C, doli incapax is stronger than the presumption of innocence. But the diachronic approach is equally natural. For instance, we should wear face masks to fight COVID-19 unless masks are proven useless: “Presumably(p), wearing a face mask protects…” is rebuttable at t1. But imagine that the virus mutates and becomes more lethal at t2. In such circumstances, “Presumably(p), wearing a face mask protects…” can become irrebuttable: even if the evidence indicated that masks were useless, wearing them would still be justified at t2 due to increased risk. Sometimes, even overwhelming evidence can be misleading, and desperate times require desperate measures. Second, the rule’s formal status determines the strength of a presumption. As Freeman (2019) explains, “[w]e may evaluate the strength not only of empirically backed warrants, but warrants backed by institutional rules.” When two presumptive rules share the same structure, the one with institutional backing seems stronger than the one without it. Consider two rebuttable rules associated with legal proceedings. First, if a person is missing for seven years, the jury must presume that she is dead unless convincing evidence indicates that she may be alive (Macagno & Walton 2012, pp. 282-283). Second, “[t] hings once proved to have existed in a particular state continue to exist in that state” (Reynolds 1897, as quoted in Macagno & Walton 2012, p. 280). So if the person was alive on Monday, the jury may presume that she was alive on Tuesday (Macagno 2010). Both rules are rebuttable and (mostly) justified on epistemic grounds, but the presumption of death is grounded in legal convention. It is the ‘presumption of law,’ so the jury must presume that “John is dead” (if the basic fact is established and defeaters are unknown). But since “Mark was alive on Tuesday” is the ‘presumption of fact,’ legal rules do not back it in the same way: the jury is authorised, but not formally required to presume “Mark was alive on Tuesday.” Accordingly, in legal context C, p is stronger than r if p is a ‘mandatory presumption’ and r is a ‘permitted presumption’ (Rescher 2006, pp. 7-8; Gama 2017, p. 567). An alternative route of synchronic analysis is also suitable for illustrating the difference. For instance, the presumption of death is weaker in a non-legal context C2 than in a legal context C1. Although a layman is justified in presuming that “John is dead” after seven years of his absence, she is not formally obliged to draw such a conclusion in non-legal deliberation. 162

On the Strength of Presumptions

Third, strength depends on a rule’s further justification. The more substantial the reasons backing the rule, the stronger the presumption that follows from it. In the COVID-19 case, the presumptive rule is the following: If it is uncertain whether wearing a mask protects other people from contracting COVID-19 [basic fact], then, all else being equal [non-defeater clause], the deliberating agent (Emma) should act as if they do [Presumably(p), wearing a face mask protects other people from contracting COVID-19]. In the Umbrella case (Bodlović 2021), Anne follows the following rule: If rain is uncertain [basic fact], then, all else being equal [non-defeater clause], the deliberating agent (Anne) should act as if “It will rain,” i.e., bring an umbrella [Presumably(p), it will rain]. Although both rules are grounded in precaution and the policy of avoiding costly errors (Bodlović 2021, pp. 292-296), the first one is stronger due to the quality of its normative foundations (Godden & Walton 2007, p. 313). What exactly does this mean? According to Godden, “the strength of presumptions can be set and variegated according to the importance of the non-alethic discursive goals warranting them” (2017, p. 508). Bringing an umbrella avoids getting soaked, whereas wearing a mask reduces the risk of spreading a dangerous disease. In conditions of uncertainty, protecting people’s lives is more important than staying dry, so Emma’s rule promotes a higher value and its justification is stronger. Consequently, “Presumably(p), wearing a face mask protects…” is stronger than “Presumably(p), it will rain.” The strength of a practical presumption depends on “what is at stake” (Ullmann-Margalit & Margalit 1992, p. 172), that is on the “inherent seriousness of the contention at issue” (Rescher 2006, p. 17).

5

By contrast, cognitive rules typically become stronger by making progress with regard to a fixed epistemic goal (truth, plausible belief).164 For instance, when Henry sees a man (walking a dog) who looks like Tom, he relies on the following cognitive rule: If visual perception indicates that Tom is in town [basic fact], then, all else being equal [non-defeater clause], the cognitive agent (Henry) should take it as true that “Tom is in town” [Presumably(c), Tom is in town]. Next, imagine that Jane tells Mary, in a similar epistemic context, that “Tom is in town.” In this case, Mary follows the rule: If some testifier (Jane) declares that Tom is in town [basic fact], then, all else being equal [non-defeater clause], the cognitive agent (Mary) should take it as true that “Tom is in town” [Presumably(c), Tom is in town]. 164 

To understand this intuition, its relationship to justifications of practical presumptions, as well as its potential

limitations, see Bodlović (2021, pp. 311-313).

163

CHAPTER 5

Both rules generate cognitive presumptions, but the first seems stronger. If resolving the issue does not require expert knowledge, then first-hand observation is more trustworthy than testimony. To see this, suppose that two sources generate contradictory beliefs: Jane remarks that Tom left town, but you happen to see Tom with your own eyes. Clearly, in such a case, treating perception as the more reliable source and concluding “Presumably(c), John is in town” is a reasonable choice. Usually, then, direct perception makes p stronger than a person’s testimony. Similarly, an expert consensus renders “Presumably(c), X happened on 13 July 1976” stronger than a hazy childhood memory. Thus, in principle, the more reliable the epistemic source, the stronger the connection between the basic fact and the presumption (Rescher 2006; Freeman 2005); the stronger the connection, the stronger the presumptive rule; and, finally, the stronger the rule, the stronger the presumption following from it. Legal theorists acknowledge that the connection’s tightness affects the presumption’s strength. Archbold (1831) distinguishes between ‘violent,’ ‘probable,’ and ‘light’ presumptions. For instance, if one sees John carrying stolen goods immediately after a robbery, in the vicinity of where it happened, “John stole the goods” becomes a violent presumption: the basic fact is a sure sign of the truth of the presumption. But if one were to find the stolen goods in John’s apartment some days after the robbery, this fact is only a plausible indicator of its truth. Consequently, “Presumably(c), John stole the goods” becomes weaker—it is not certain but probable. Finally, if one were to find the stolen goods in John’s apartment years after the robbery, “John stole the goods” becomes a weak or light presumption (Archbold 1831, p. 114; Macagno & Walton 2012, pp. 284-285). Regarding evidential force, Rescher (2006, p. 66) distinguishes between two types of cognitive presumptions. First, some presumptions are supported by weak evidence and a complete absence of defeaters. If a non-expert were to say “This diet is healthy,” and defeaters were unknown, it might be reasonable to trust her testimony. Second, some presumptions are backed by solid evidence but, at the same time, challenged by insufficient defeaters. For instance, one may conclude “Presumably(c), the vaccine against COVID-19 is safe” based on prevailing expert opinion (i.e., the preponderance of evidence) although an expert minority contends “The vaccine is unsafe.” Although the previous analysis was conducted in a synchronic fashion, the diachronic approach is also appropriate: a presumptive rule might become stronger over time. If values or stakes are more serious at t2 than at t1, the practical rule becomes stronger at t2. Also, if some evidence discovered after t1 tightens the connection between a fact and a presumption, the cognitive rule becomes stronger at t2. The overall point is simple: a rule’s strength might vary even when its general structure and formal status remain unchanged. I explained this by focusing on the rule’s objective justification, but one could also consider its rhetorical effectiveness, i.e., the extent to which the rule coheres with the views of an intended audience (see Section 3.1).

164

On the Strength of Presumptions

Finally, the strength of a presumption depends on the absence of defeaters. According to Godden & Walton (2007, p. 337), the “conditions for defeating the presumptive status … form part of the presumption rule.” Thus, the strengths of the rule “If q, then—all else being equal—‘Presumably, p’” and the conclusion “Presumably, p” depend on the content of a non-defeater clause. The fewer (more) defeating options the Challenger has, the stronger (weaker) the presumptive rule.165 First, if a rule is a formal convention in C1, the Challenger cannot defeat it by using a tenability criticism. For instance, in criminal law, “The accused is presumed innocent until proven guilty” is an institutional convention, so its acceptability is beyond question. By contrast, criticising the rule is usually allowed in a non-legal context C2. Suppose Peter accuses John of stealing his pencil but presumes John’s innocence due to insufficient evidence to the contrary. Peter’s co-worker Chris can challenge such reasoning by saying: Chris: “We should not go easy on potential thieves and encourage stealing. The rule ‘Innocent, until proven guilty’ is wrong. You should consider John guilty until he proves otherwise.” [rule tenability criticism] Since the tenability criticism is permitted in C2 and prohibited in C1, Chris has more attacking options and an easier job defeating the presumptive rule. Consequently, the rule and the final presumption seem weaker in C2. Second, in some circumstances, the Challenger might be unable to ‘undercut’ a presumption.166 Consider the Cat case. Andy infers “Presumably(c), Andy’s cat is in the tree” from “Andy sees that a cat, which looks exactly like his cat, is in the tree” (Bodlović 2021). Suppose that Jane tries to undercut his reasoning:

5

Jane: Andy, the neighbour has a cat that looks exactly like yours.

[undercutting defeater]

Jane’s remark forces Andy to suspend judgment based on prior likelihoods. How likely is it that, on an ordinary day, Andy’s cat will be in that particular tree at that time? If “Andy’s cat is in the tree” and “Andy’s cat is not in the tree” have similar likelihoods in C1, the cognitive presumption is defeated in C1, and Andy must suspend judgment. But if “Andy’s cat is in the tree” is sufficiently more likely in C2, it remains a cognitive presumption in C2 (Bodlović 2021, pp. 313-315). Since the undercutter is ineffective in C2, challenging the presumptive rule becomes more demanding. As a result, the presumption is stronger in C2 than in C1. 165  For an overview of various defeaters that cancel a presumption and that, for the most part, constitute the presumptive rule’s ‘non-defeater clause,’ see Prakken (2004, pp.41-42), Rescher (2006, p. 35), Godden & Walton (2007, pp. 338-341); Godden (2011, p. 3; 2017, pp. 506–507), Bodlović (2019b, pp. 587-588; 2021, pp. 296-297), etc. 166  The so-called undercutting defeater is a piece of evidence that renders the evidential connection between premise q and conclusion p unreliable. Typically, an undercutter forces an epistemic agent to suspend judgment or remain agnostic towards conclusion p (Pollock 1987).

165

CHAPTER 5

Third, features of a particular presumptive rule might limit the Challenger’s options for defeating it. Usually, the Challenger has (at least) two ways of questioning the rule’s applicability. First, she can criticise the premise, i.e., refute the rule’s antecedent (prove non-q). Second, she can rebut the presumption, i.e., deny the rule’s consequent (prove non-p). For illustration, suppose the Challenger wants to defeat an argument from an expert opinion: Since a famous virologist said that the vaccine against COVID-19 is safe [basic fact], all else being equal [non-defeater clause], “Presumably(c), the vaccine against COVID-19 is safe.” The premise tenability criticism differs from presumption rebuttal: showing that the virologist did not say “The vaccine is safe” does not prove “The vaccine is unsafe,” and vice versa. For instance, by proving that the journalist misinterpreted the virologist’s statement, the Challenger does not show that the vaccine has dangerous side effects. Also, by proving dangerous side effects, the Challenger does not disprove the basic fact—it remains entirely unchallenged that the virologist said this or that. Thus, the two strategies are logically independent. But suppose that an agent reasons as follows (Bodlović 2019b, p. 589): Since it is not known that the water is contaminated [basic fact/ignorance premise], all else being equal [non-defeater clause], “Presumably(c), the water is not contaminated.” In this case, refuting the premise entails rebutting the presumption. That is, the Challenger must show that some contamination has been found to deny “It is not known that the water is contaminated.” But arguing that the water is contaminated directly rebuts “Presumably(c), the water is not contaminated.” Thus, the basic fact criticism and presumption rebuttal cannot be truly distinguished, and the Challenger has one genuine attacking option (instead of two). Due to a lack of attacking diversity, defeating conclusion p of an argument from ignorance might be more challenging than defeating conclusion r of an argument from expert opinion.167 In this sense, p might be stronger than r. Fourth, even if the Challenger could use all defeaters, and they were genuinely diverse and potentially effective, instances of a particular defeater can be falsified and, thus, unavailable. Consider the Polished rice case, discussed in Freeman (2019). Suppose that, by eating polished rice, chickens normally develop fatal polyneuritis. Given this finding, an agent can make an inference to the best explanation: Since the chickens ate polished rice and developed fatal polyneuritis [complex basic fact], all else being equal [non-defeater clause], “Presumably(c), chickens died because they ate polished rice.” To rebut the presumption, the Challenger should offer a more plausible alternative explanation. As Freeman (2019) remarks, she might argue that chickens were genetically predisposed to develop polyneuritis, or discover that an environmental condition or some 167 

For a detailed discussion on arguments from ignorance and defeaters, particularly with regard to practical and

cognitive presumptions, see Bodlović (2019b, pp. 594-597).

166

On the Strength of Presumptions

other dietary ingredient caused it. Crucially, the more rival explanations are falsified, the fewer rebutting options the Challenger has, and, consequently, the stronger the initial ‘polished rice hypothesis.’ While commenting on Balfour’s A Defence of Philosophic Doubt, Giddings elaborates the strength of presumptions in a similar way: Philosophy is concerned with the grounds of presumption, and with ultimate presumptions. Its business is to bring our assumptions, beliefs and faiths face to face, and let those survive that can. The survivors we may not accept as certainties, but we may accept them as presumptions. The strength of presumption increases as the death-rate of beliefs rises. (Giddings 1922, p. 623) In summary, strength depends on various contextual factors, the nature of foundations, and the evidence that potentially defeats a presumption. The more (fewer) defeaters are potentially available, the weaker (stronger) the presumption. A theorist might specify defeaters in a non-defeater clause, thereby making them a constitutive part of a presumptive rule.

3.3 Deontic Factors168 Although different types of defeaters may influence p’s strength, scholars typically explore conditions of proving non-p. On Ullmann-Margalit’s view, “an index of the strength of the presumption” is correlated with “the weight of the reasons [...] required for the rebuttal” (1983, p. 152). Accordingly, a presumption “based on very weak grounds should be easily overturned” (Godden & Walton 2007, pp. 337-338), and a “presumption is strong whenever there is a need for exceptionally weighty evidence in order to change the presumption” (Räikkä 1997, p. 471).

5

Presumptions supposedly allocate the burden of proof to those who challenge them, so explaining their force in deontic fashion—in terms of dialectical obligations—is appropriate: the more (less) demanding the Challenger’s obligation to prove non-p, the stronger (weaker) “Presumably(p), p” is. [P]resumptions will differ in strength according to just how much it takes to defeat them—that is, on just how great a burden of proof is being carried by anyone seeking to defeat them. Presumptions are strong or weak as their associated burden is heavy or light. (Rescher 2006, p. 18) In line with the synchronic approach, Rescher maintains that the ‘presumption of innocence’ is stronger than the legal ‘presumption of sanity’ (2006, p. 18).169 The Challenger must present a conclusive argument to rebut the former, i.e., show that “The accused is guilty” is proven ‘beyond a reasonable doubt.’ However, to rebut the latter, the Challenger must prove that “The accused was insane” is (at least slightly) more plausible 168 

Section 3.3 is based on Bodlović (2019a).

169  According to this legal presumption, we shall presume that the accused was sane (in the right state of mind) when the crime in question was committed.

167

CHAPTER 5

than “The accused was sane.” Thus, meeting the ‘preponderance of evidence’ standard suffices to rebut the presumption of sanity, but not the stronger presumption of innocence. The strength of “Presumably(p), p” depends on the weight of the rebuttal, which primarily hinges on the ‘standard of proof’ that the Challenger must satisfy to defend non-p successfully. Finally, the “standard of rebuttal” (Godden & Walton 2007, p. 314) is determined by values and risks underlying a concrete piece of presumptive reasoning: The weight of the reasons, or counterevidence, required for rebuttal varies with the degrees of strength of the presumption, which in general depend on what is at stake. (Ullmann-Margalit & Margalit 1992, p. 172) If “Presumably(p), p” promotes a higher non-epistemic value than “Presumably(p), r,” the Challenger must satisfy a more demanding standard of proof to rebut p (Rescher 2006, p. 18; Godden 2017, p. 508). For instance, if proceeding on p protects lives and proceeding on r promotes politeness, the Challenger should present overwhelming evidence to rebut p, while any plausible argument might be enough to rebut r.170 In conditions of uncertainty, putting lives in danger is potentially more costly than allowing impolite behaviour. Importantly, factors other than the standard of proof affect the weight of the rebuttal, thereby making “Presumably, p” stronger or weaker. Many such factors are contextual and could have been reviewed in Section 3.1. Still, since they directly affect the weight of the Challenger’s obligation to defend non-p, I study them from a deontic perspective. First, the strength of “Presumably, p” varies depending on time restrictions. The burden is heavier if the Challenger must demonstrate non-p (“Wearing a face mask does not protect other people from contracting COVID-19”) within minutes rather than weeks. Second, the strength of “Presumably, p” depends on the accessibility of evidence. The Challenger’s job is more demanding if he cannot easily access evidence for non-p. For instance, the evidence supporting non-p may belong to the distant future or past (Walton 2014, p. 1; Aijaz et al. 2013, pp. 266-267), or acquiring it may involve a morally impermissible action (Ullmann-Margalit 2000, p. 74). Third, the strength of “Presumably, p” depends on whether the evidence for non-p is publicly transparent. For instance, if the proof for “The vaccine against COVID-19 is not safe” is novel and entirely unknown to the presumption’s User (and the general public), the Challenger incurs a demanding presentational burden. She will need to present every piece of evidence and elaborate on it in a detailed manner. 170 

Legal and argumentation theorists distinguish between four standards of proof. The Challenger satisfies the weakest

(1) scintilla of evidence (SE) standard by (a) supporting non-p with an undefeated argument (however weak this argument may be). Moreover, she satisfies the moderate (2) preponderance of evidence (PE) standard if she satisfies SE, and (b) her argument for non-p is more plausible than the strongest argument for p. Furthermore, the Challenger satisfies the strong (3) clear and convincing evidence (CCE) standard if she satisfies PE, (c) the weight of her argument for non-p exceeds some

threshold α, and (d) the difference between the weight of her argument for non-p and of the strongest argument for p exceeds

some threshold β. Finally, she satisfies the highest (4) beyond a reasonable doubt (BRD) standard if she fulfils CCE, and (e)

the weight of the strongest argument for p does not exceed some threshold γ (Macagno & Walton 2012, p. 275).

168

On the Strength of Presumptions

The strength of both practical and cognitive presumptions depends on time constraints and the accessibility of evidence. But, arguably, only the strength of practical presumption can be explained by referring to differences in the standard of proof. Whether truth-directed, epistemic contexts—and cognitive presumptions belonging to such contexts—permit different standards of proof is controversial (Hahn & Oaksford 2007; Dare & Kingsbury 2008; Koplin & Selgelid 2015). Hence, sometimes, when time and accessibility are not an issue, the strength of cognitive presumption is not correlated to the weight of rebuttal. I show this by means of diachronic analysis: cognitive presumption p can be weaker at t2, although proving non-p is not easier at t2 than at t1. Practical and cognitive presumptions thus entail distinct conceptions of strength.

4 Cognitive Presumptions and the Weight of Rebuttal171 Justificatory and deontic factors are closely connected as far as the strength of practical presumptions is concerned: “the appropriate standard of rebuttal will depend on the foundational grounding of the presumption itself” (Godden & Walton 2007, p. 315). The higher the stakes justifying a presumptive rule, the weightier the Challenger’s obligation to prove non-p. By contrast, as far as cognitive presumption is concerned, justificatory and deontic factors seem logically independent. “Presumably(c), p” can become less justified and, thereby, weaker without making the Challenger’s burden of rebuttal any lighter.

5

According to Rescher, “a [cognitive] presumption is a provisional estimate that represents our most plausible candidate for truth” (2006, p. 71). Since plausibility comes in degrees, the strength of “Presumably(c), p” correlates with the degree of p’s plausibility. Therefore, it follows: If p becomes less plausible (at t2), “Presumably(c), p” becomes weaker (at t2). Typical cognitive presumptions become less plausible by losing evidential support. To show this, let’s substitute ‘p becomes less plausible’ with ‘p lost some evidential support.’ If p lost some evidential support (before t2), “Presumably(c), p” becomes weaker (at t2). This is true. But what happens if we interpret “‘Presumably(c), p’ becomes weaker” in deontic terms? As we have seen, practical presumption p becomes weaker when proving non-p becomes more difficult. So, let’s apply this model of strength to cognitive presumption p. If p lost some evidential support (before t2), it becomes easier to prove non-p (at t2). This is wrong. It is easy to think of an example that satisfies the antecedent but not the consequent. For instance, suppose that q is one among many reasons that support “Presumably(c), p,” and that u undercuts the reliability of the connection between q and “Presumably(c), p” at t1. Although “Presumably(c), p” lost some evidential support and 171 

Section 4 is almost entirely based on Bodlović (2019a).

169

CHAPTER 5

became weaker at t2, proving non-p is not easier at t2. Since undercutters do not directly support negations, u does not make the Challenger’s obligation to defend non-p any weaker.172 Consider the following example. After twenty years, Henry returns to his home town and sits in a bar. At t0, he sees a man walking a dog and knows that this man is either Tom or his twin brother Frank (Henry’s long-lost high-school friends). By supposition, there are only two plausible truth candidates: “Tom is walking a dog” and “Frank is walking a dog.” With the knowledge that the brothers are twins, Henry’s perception makes the two propositions equally plausible, so neither “Tom is walking a dog” nor “Frank is walking a dog” is a cognitive presumption at t0. At t1, Henry overhears the bartender saying: “Look at that dog lover Tom over there!” Since, normally, one should “accept at face value the declarations of other people” (Rescher 2006, p. 31), “Tom is walking a dog” becomes the most plausible proposition, i.e., a cognitive presumption at t1. Note that Henry’s reasoning rests on a complex basic fact because the bartender asserts two statements: “Tom is walking a dog” and “Tom is a dog lover.” In my view, these statements provide independent, convergent support for “Presumably(c), Tom is walking a dog.”173 But at t2, someone says: “Frank is a dog lover, too.” At that moment, “Tom is a dog lover” (from t1) ceases to be a reliable indicator that Henry is looking at Tom and not at Frank: the person walking a dog is either Tom or Frank, twin brothers look identical, and Frank loves dogs as much as Tom. In effect, “Tom is a dog lover” provides weaker support for “Presumably(c), Tom is walking a dog” at t2 than at t1. The following is crucial: although “Frank is a dog lover” undercuts one line of argument and weakens “Presumably(c), Tom is walking a dog,” it does not make “Frank is walking a dog” any easier to defend at t2. Imagine that the Challenger initially believes “Frank is walking a dog” based on visual appearance. At t0, the information about Frank’s twin brother undercuts the Challenger’s inference from a visual impression to “Frank is walking a dog.” At t1, the bartender’s statement “Tom is walking a dog” rebuts the Challenger’s belief. Crucially, at t2, the statement “Frank is a dog lover” does not make the Challenger’s obligation to prove “Frank is walking a dog” any weaker. “Frank is a dog lover” does not support “Frank is walking a dog” since the Challenger knows from t1 that Tom is a dog lover too. Instead, at t2, “Frank is a dog lover” shows that—as far as dog preferences are concerned—the Challenger should 172  An undercutting defeater undermines the connection between a premise q and a conclusion p, but this does not entail that non-p is true. As Pollock (1987, p. 485) illustrates, an undercutter “X is illuminated by red lights” (u) is consistent both with “X looks red to me” (q) and “X is red” (p). So, it does not rebut “X is red” but simply makes it evidentially uncertain. Instead of instructing an epistemic agent to adopt the contrary belief, an undercutter normally instructs her to suspend judgment on a matter: as long as X is illuminated by red lights, she cannot know whether X is red, white, or, perhaps, yellow. 173 

Since Henry has lost contact with his friends over the years, he might reasonably expect that either Tom or Frank

live in another neighbourhood, or perhaps even in another town. Thus, the fact that the local bartender recognised Tom, and not Frank, must count as evidence for “Tom is walking a dog.” For all that Henry knows, Frank could live in another state. Also, assuming that Henry doesn’t know anything about Tom’s or Frank’s preferences for dogs, the bartender’s claim “Tom is a dog lover” must count as evidence for “Tom is walking a dog.” For all that Henry knows, Frank might hate dogs.

170

On the Strength of Presumptions

suspend judgment on who, in fact, is walking a dog. The reason to suspend judgment on non-p hardly brings the Challenger any closer to proving non-p.174 Hence, while the evidence from t2 weakens “Presumably(c), p,” it does not directly contribute to establishing non-p and, therefore, does not make the Challenger’s burden of rebuttal any lighter. To sum up, it sometimes happens that one cannot entirely explain the strength of cognitive presumption in view of the rebuttal’s weight. The force of epistemic justification and the degree of plausibility do not always correlate with deontic factors. When the strength of practical and cognitive presumptions is concerned, different mechanisms seem to be in play.

5 Conclusion In this chapter, I systematically reviewed the various directions taken in studying the strength of presumptions. Conducting so-called ‘conceptual research’ entails explaining the meaning of “Presumably, p” in terms of p’s epistemic strength or pragmatic (illocutionary) force. Conducting so-called ‘strength research’ entails exploring various factors that make “Presumably, p” stronger or weaker. Strength research may include the ‘synchronic approach’ and the ‘diachronic approach.’ The former seeks to explain what makes presumption p stronger or weaker at some fixed time t. The latter seeks to explain what makes p stronger or weaker over time (e.g., why is p stronger at t2 than at t1).

5

I identified three categories of factors affecting the strength of presumptions. First, there are ‘contextual factors,’ such as the dialogue’s collective goal, cooperativeness, and the features of a particular audience. Second, there are ‘justificatory factors.’ Presumption’s strength, for instance, depends on whether the presumptive rule is conclusive or defeasible, formal or informal, backed by strong or weak reasons, allows or disallows many defeaters. Third, ‘deontic factors’ determine how difficult it is for the Challenger to satisfy her dialectical obligation, i.e., to rebut “Presumably, p.” The more difficult it is to prove non-p—due to a demanding standard of proof, inaccessible evidence, or time constraints—the stronger the presumption p. The aforementioned categories are intertwined. For instance, local contextual factors (such as practical stakes and background information) may affect the structure and content of p’s justification. Also, factors justifying practical presumptions directly affect deontic factors: the higher the value supporting the presumptive rule, the more weighty the Challenger’s burden of proof. Apart from identifying the different research directions and factors determining the strength of presumptions, I argued that the relationship between justificatory and deontic factors depends on a type of presumption. On the standard view, if p is a practical presumption, the weaker the non-epistemic reasons justifying p, the easier the Challenger proves 174  Generally, in the realm of plausible reasoning, that p becomes less plausible does not automatically entail that non-p becomes more plausible (see Woods, Irvine & Walton 2004, p. 299).

171

CHAPTER 5

non-p. But sometimes, this does not apply to cognitive presumptions: from a diachronic perspective, even if p’s epistemic justification is weakened at t2, proving non-p can remain equally demanding (like at t1). Thus, not only do practical and cognitive presumptions have distinct contextual, pragmatic, dialogical, and deontic functions (see Bodlović 2020b; 2021; forthcoming), but they are also underpinned by distinct conceptions of strength. Acknowledgments: An early version of this chapter was presented at the 9th ISSA Conference (Amsterdam, July 3-6 2018), and published in the Proceedings of the Ninth Conference of the International Society for the Study of Argumentation (eds. B. Garssen, D. Godden, G. R. Mitchell, and J. H. M. Wagemans, 106-117. Amsterdam: Sic Sat), under the title “What determines the strength of practical and cognitive presumptions?” I thank the organisers of this conference, the participants, and the editors for their valuable comments and suggestions. Also, I thank my supervisors, Jan Albert van Laar and Barteld Kooi, for their helpful advice.

172

On the Strength of Presumptions

5

173

6

CHAPTER 6 Bringing It All Together: Presumption Models, and Conceptual Consequences

CHAPTER 6

Writing long books is a laborious and impoverishing act of foolishness… A better procedure is to pretend that those books already exist and to offer a summary. —Borges, The Garden of Forking Paths

176

Bringing It All Together: Presumption Models, and Conceptual Consequences

Bringing It All Together: Presumption Models, and Conceptual Consequences Abstract: In this chapter, I summarise the results of previous chapters. On the one hand, we use practical presumptions when we must make decisions under time constraints, but lack epistemic grounds that would favour any particular course of action. On the other hand, we use cognitive presumptions when we accept the most epistemically plausible proposition at face value. First, I construct and present functional models of practical and cognitive presumptions in terms of their (1) basic characterisations, (2) logical functions, (3) pragmatic functions (justifications), (4) deontic functions, (5) dialogical functions, (6) conditions of defeat, and (7) conceptions of strength. Second, I discuss conceptual consequences arising from relevant differences between practical and cognitive presumptions. Insofar as ‘presumption’ as an “umbrella” term is concerned, I argue in favour of the so-called ‘moderate pluralism.’ Keywords: cognitive presumptions, functions, models, pluralism, practical presumptions. In everyday life, we must often make decisions under time constraints, but without available evidence that would favour any particular course of action. If chances of rain are uncertain when we are about to leave our apartment, should we bring an umbrella? If a person whose presence will ruin our evening may attend a party, should we go? Practical presumptions promote precaution: in conditions of uncertainty, when it is time to decide, we should avoid greater harm unless greater harm is proven unlikely.175 Simply put, since the inconvenience of bringing an umbrella is less harmful than getting soaked (and potentially becoming ill), we should bring an umbrella unless we know that it won’t rain. Thus, in the context of uncertainty and under time pressure, proceeding on “Presumably, it will rain” is a reasonable practical policy.

6

While practical presumptions are common in practical deliberation, cognitive presumptions are unavoidable in theoretical reasoning. Whenever we accept the most plausible proposition at face value, we concede a cognitive presumption. In normal circumstances, we accept propositions like “Presumably, the cat is in the tree” based on 175  There are many formulations of the ‘precautionary principle,’ but theorists agree that “[c]entral to all precautionary approaches is the notion of uncertainty” (Allhoff 2014, p. 119). The notion of epistemic uncertainty, however, is ambiguous. On the one hand, p may be uncertain because p’s probability is not sufficiently precise (“The chance of rain is 30-60%”). On the other hand, p may be uncertain because p’s probability is entirely unknown (“The chance of rain has not been estimated”) (see Allhoff 2014, p. 110). In this dissertation, I regarded p as an epistemically uncertain proposition if the reasonable agent cannot tell—for whatever reason—whether p is true or false (given p’s justification and a contextually appropriate standard of proof). On this account, p can be uncertain even if its probability is known and exact: if we know that the probability of rain is exactly 50%, it still appears reasonable to regard “It will rain” as an uncertain proposition. To my mind, this broader notion of uncertainty does not conflict with how precautionary principles are supposed to operate.

177

CHAPTER 6

visual observation, “Presumably, the Earth is round” based on scientific evidence, and “Presumably, this street leads to the city centre” based on the testimony of passers-by, unless we obtain proof that these propositions are false/unjustified. In a nutshell, trusting reliable sources at the cost of acquiring some false/unjustified beliefs is less harmful than repeatedly suspending judgment in order to avoid false beliefs. Radical scepticism comes at a high price in everyday cognition, so sticking with cognitive presumptions seems like a reasonable epistemic policy. How do argumentation theorists model presumptions? What are their dialogical functions? What does their justification look like? Which dialogical patterns do they create? Do they shift the burden of proof? How are they defeated? What determines their strength? In this dissertation, I sought to reconstruct, evaluate, and improve the answers that traditional, judicially inspired accounts of presumption offer in response to these questions. I build upon these theories, and develop an approach with its own characteristic features. Section 1 and Section 2 summarise the dissertation’s dialectical results by systematically presenting the models of practical and cognitive presumptions. Section 3 comments on the linguistic and conceptual issues resulting from the dissertation’s typological conclusions, regarding the many differences between practical and cognitive presumptions.

1 Practical Presumption (PP) 1.1 Basic Characterisation of PP Practical presumptions are propositions with a modifier (qualifier) ‘presumably.’ “Presumably, p” indicates that some proposition p is acceptable unless proven otherwise. In a persuasion dialogue, “Presumably, p” would entail that the proponent is not obliged to defend her commitment p (in the concluding stage), whereas the opponent must give sufficient argument against p’s acceptability (before the concluding stage). In a practical case, “Presumably(p), p” expresses an action-oriented commitment: a proposition one proceeds on due to some non-epistemic reason (such as safety, health, protection of rights, etc.). So, despite its epistemic defeasibility, a practical presumption represents a non-epistemic commitment. The dissertation did not explore the ontological properties of presumptions. Instead, it sought to develop a functional account, and, as far as an ontological description is concerned, treated presumption as a qualified proposition (see Ullmann-Margalit 1983; Hansen 2003; Rescher 2006; Walton 2014; Godden 2017, 2019).176 Still, we might 176 

Arguably, many presumptions’ functional properties—e.g., how they influence deliberation or inquiry; how they

affect the burden of proof, etc.—can be discussed independently of whether we describe a presumption as a proposition, a belief, a commitment, a qualifier, a relation, a process, or a speech act. On the one hand, such descriptions influence how we speak about presumptions, and, on the other hand, they are relevant for a more fundamental, ontological, inquiry of the nature of presumptions. However, neither linguistic nor ontological features were this dissertation’s primary concern.

178

Bringing It All Together: Presumption Models, and Conceptual Consequences

distinguish between four general aspects of a practical presumptive status, conveyed by the modifier ‘presumably(p):’ a procedural, a justificatory, a modal, and a deontic aspect.

6

Table 1: Practical presumption: four aspects of the ‘presumably(p)’ modifier (the Umbrella case)

1.2 The Logical Function of PP In logical terms, practical presumptions are portrayed as conclusions of presumptive reasoning. Although scholars sometimes equate presumptions and presumptive rules, this research treated them as conclusions inferred from basic facts. Below, I provide a diagram of the core structure of presumptive practical reasoning by using the Umbrella case.

179

CHAPTER 6

Diagram 1: Presumptive practical reasoning: the core structure (the Umbrella case)

Whether presumption is modelled as a conclusion or a rule of reasoning is not essential for developing a theory of presumption. Regardless of what position the theorist upholds, she will face an identical set of theoretical questions. In either case, she will need to explore the rule’s structure, its justification, and the consequence of its application (the conclusion “Presumably, p”). Hence, I agree with Walton that picking between a ‘conclusion-based’ and a ‘rule-based’ model of presumption depends, for the most part, on the “choice of which language to adopt” (2014, p. 115).

1.3 Justification and the Pragmatic Function of PP Practical presumptions are closely related to the argument from ignorance. For instance, the legal presumption of innocence is reducible to the argument from ignorance, since its basic fact corresponds to an ad ignorantiam’s ignorance premise. That is, the ignorance premise “The prosecutor failed to prove that ‘The accused is guilty’” implies that, ultimately, “The accused is guilty” remains uncertain, and, as a result, the accused must be presumed innocent and set free. In the Umbrella case, however, the basic fact (associated with uncertainty) entails the ignorance premise: “There is 50% chance of rain” implies that “It is not proven that it will not rain.”177 The latter ignorance premise, by means of ad ignorantiam, supports “Presumably(p), it will rain” and bringing an umbrella. Our next question is: What justifies a presumptive rule? Why should Anne bring an umbrella if rain is uncertain? Basically, her endorsement of a presumption is motivated by a significant chance of making a mistake. If Anne brings an umbrella, there is a 50% chance of needlessly carrying it around. If she does not bring an umbrella, there is a 50% chance of getting soaked and risking illness. When mistakes are likely and time for 177

In the Umbrella case, we can understand “There is 50% chance of rain” and “Rain is uncertain” as the following

conjunction: “It is not proven that it will rain, and it is not proven that it will not rain.”

180

Bringing It All Together: Presumption Models, and Conceptual Consequences

deliberation is limited, avoiding costlier errors is a recommended policy. Anne estimates that getting soaked is potentially costlier than needlessly carrying an umbrella because she values health more than convenience. Consequently, she decides to err on the side of safety and follow the presumptive rule “If rain is uncertain, then, all else being equal, proceed on ‘It will rain’.” On this view, the so-called pragmatic function of practical presumption helps to avoid greater harm in the context of deliberation pressure and genuine (plausibilistic) uncertainty. Below, I provide a diagram of the complete structure of presumptive practical reasoning by using the Umbrella case.

6

181

Diagram 2: Presumptive practical reasoning: the complete structure (the Umbrella case)

CHAPTER 6

182

Bringing It All Together: Presumption Models, and Conceptual Consequences

1.4 Patterns of Interaction and the Deontic Function of PP How does a practical presumption influence argumentative interaction? Is the presumption’s opponent (immediately) obliged to defend his rejection, and is the presumption’s proponent exempted from taking up the burden of proof? In principle, a practical presumption imposes the burden of reasoning on its opponent: after rejecting “Presumably(p), p” at t2, he must either explain or argue for his view at t3. Since explaining does not aim at persuasion, the opponent can present any reason underlying his view. For instance, Jim may elucidate his denial of “Presumably(p), it will rain” by saying: “The tea leaves tell me that it will not rain.” Although this explanation is unacceptable to Anne and thereby dialectically inadequate, explanations should not, in principle, be banned by dialectical rules. However, due to uncertainty and time constraints, Jim is strongly advised to undertake the burden of proof and provide arguments consisting of adequate, relevant, and strong reasons (such as “According to the weather forecast, the certainty of rain is only 25%). Persuasive arguments do not waste precious time and may actually affect uncertainty. I illustrate the opponent’s permitted, yet suboptimal response with a curved line in the profile below.178

6

Figure 1: Practical presumption: a profile of a dialogue (the opponent Jim / the Umbrella case) 178

In the following schemes and diagrams, the colour green (e.g., green box) indicates a permitted move that is entirely

legitimate—usually optimal—in the dialogical conditions at hand. Likewise, the colour orange (e.g., orange box) signals a permitted move that, although suboptimal, is quite reasonable in the conditions at hand. Finally, the colour red (e.g., red box) indicates a contribution that, although permitted in principle, is typically unconstructive or is even destructive, in particular dialogical circumstances.

183

CHAPTER 6

By deciding to argue at t3, Jim assumes the burden of production: in order to avoid unsupported claims and do what is necessary to win the debate, Jim must produce an adequate and relevant argument. Although offering a convincing argument is optimal, Jim is also permitted, at t3 and at the beginning of the argumentation stage, to provide some minor argument to get things going (e.g., “We should not bring an umbrella because, according to the weather forecast, the rain is only 45% likely”). However, Jim also assumes the burden of persuasion. This burden is allocated in the opening stage, influences the interaction in the argumentation stage, and is evaluated in the concluding stage. First, the burden of persuasion specifies the standard of proof that Jim’s arguments must meet in order to successfully justify “We should not bring an umbrella”. For instance, Jim may be required to prove that rain is less than 40% likely. Second, the burden of persuasion implies that the party who carries it loses the discussion if her arguments do not meet the specified standard. So, if Jim is able to show that rain is only 41% likely, parties should still bring an umbrella. Due to precaution, in the concluding stage, any remaining uncertainty counts in Anne’s favour. Thus, the proponent of a practical presumption is exempted from the burden of persuasion, but does she take up the burden of production? First, this depends on the transparency of a presumption’s justification. Suppose parties agree that, when rain is uncertain, precaution represents a reasonable policy, and suppose they both notice that the sky is cloudy. In such circumstances, Anne does not seem obliged to present the obvious justification for “Presumably(p), it will rain.” Second, her burden of production also depends on the strength of counterevidence. For instance, if Jim offers a persuasive counterargument at t3—such as, “According to the weather forecast, the rain is only 25% certain”—intuitively, Anne must support her global presumption at t4. Accordingly, practical presumption entails a binary asymmetry when its justification is transparent and the opponent’s argument is weak. Otherwise, it involves a so-called degree asymmetry: although both parties must support their views, the presumption’s opponent must meet the higher standard of proof in the concluding stage. In the profile below (Figure 2), I illustrate the interaction pattern where the presumption’s proponent takes up the burden of production.

184

Bringing It All Together: Presumption Models, and Conceptual Consequences

Figure 2: Practical presumption: a profile of a dialogue (the proponent Anne / the Umbrella case)

6

Bearing this in mind, we can formulate the so-called deontic function of practical presumption as follows: Production of arguments (argumentation stage) a) The proponent advancing “Presumably(p), p” does not shoulder the burden of production. o If reasons for “Presumably(p), p” are entirely transparent. o If reasons for “Reject: ‘Presumably(p), p’” are minor/weak. b) The opponent rejecting “Presumably(p), p” is obliged to provide either an explanation or an argument, but is strongly advised to provide an argument and discharge the burden of production. Evaluation of arguments (concluding stage) a) The proponent advancing “Presumably(p), p” does not shoulder the burden of persuasion. b) The opponent rejecting “Presumably(p), p” shoulders the burden of persuasion. 185

CHAPTER 6

Although practical presumptions might entail other asymmetries in taking up burdens, the asymmetries concerning evidence accessibility or commitment strictness are more typically associated with cognitive presumptions.

1.5 Dialogical Functions of PP If practical presumptions asymmetrically allocate the burden of persuasion, how does this benefit the dialectical procedure? I distinguish between two dialogical functions of presumptions. First, since practical presumptions are employed in deliberative dialogues, they help in deciding on the course of action (and not, e.g., in acquiring knowledge or justified belief). I label this obvious dialogical role the ultimate contextual function of practical presumption. Second, since practical presumptions asymmetrically allocate the burden of persuasion, they enable dialogical progress in circumstances of pressure and uncertainty. Suppose that Anne and Jim have exhausted all evidential resources during the argumentation stage and now must proceed to the concluding stage—there is simply no more time to deliberate. If rain remains uncertain, their deliberation will become stuck. In such circumstances, practical presumptions break the deadlock. Since Jim shoulders the burden of persuasion (deontic function), the difference of opinion is resolved in Anne’s favour. Importantly, practical presumptions promote non-arbitrary decision-making: Anne and Jim should proceed on “It will rain” and bring an umbrella because, by doing so, they avoid more significant harm (pragmatic function). I label this paradigmatic role of enabling dialogical progress the (special) dialogical function of practical presumption.

1.6 Conditions of PP’s Defeat If a practical presumption stands until proven otherwise, how can the opponent discharge the burden of proof? How can Jim defeat “Presumably(p), it will rain?” There are three lines of attack. First, the opponent may attack “Presumably(p), p” by arguing that non-p, i.e., rebutting p. Second, he might challenge one of the premises of presumptive practical reasoning (see Diagram 2) and thereby engage in premise tenability criticism (e.g., overriding). Finally, the opponent may argue that in presumptive practical reasoning, some acceptable premise q does not warrant some conclusion r, thereby focusing on the connection criticism (e.g., undermining, undercutting). However, this dissertation explored a related question that has mostly escaped the attention of argumentation scholars. Instead of analysing what types of evidence defeat practical presumptions, it investigated when, and in what circumstances, a particular type of connection criticism—the so-called undercutting defeater—cancels a practical presumption. In Pollock’s (1987) classic account, the undercutter u is the piece of evidence that renders the evidential connection between premise q and conclusion r unreliable, despite 186

Bringing It All Together: Presumption Models, and Conceptual Consequences

being consistent with both q and r. For instance, Jim’s undercutter “The weather forecast website is playing an April Fool’s joke on the citizenry” (u) challenges the connection between “The weather forecast website estimates that there is 50% chance of rain” (q) and “There is 50% chance of rain” (r), despite being consistent with both propositions. Typically, an (undefeated) undercutting defeater commits an epistemic agent to adopt a sceptical stance and suspend her belief that the conclusion is true. The dissertation showed that u’s potential to defeat “Presumably(p), p” depends on p’s prior likelihood. Suppose that the chance of rain is 50% on a regular day of the season in some geographical area. Under these circumstances, an April’s Fool joke makes the weather forecast unreliable, but does not defeat “Presumably(p), it will rain.” The essential components of presumptive practical reasoning (evidential uncertainty and deliberative pressure) remain unchanged, and parties should still bring an umbrella on precautionary grounds. Table 2 summarizes how an undercutting defeater affects a practical presumption, given different prior likelihoods.179

6

179 

In a typical case, p’s presumptive status is defeated when proceeding on p becomes unreasonable because some

component of presumptive reasoning is challenged. But what happens if p (“It will rain”) becomes epistemically justified? In such a case, the essential premise of presumptive practical reasoning (indicating the genuine uncertainty of “It will rain”) becomes untenable, so, strictly speaking, the (local) presumptive status of “It will rain” must be defeated. However, proceeding on “It will rain” remains reasonable once we know that rain is likely. So, even if the presumption’s justificatory aspect is challenged, its procedural aspect may remain intact: we should still act on “It will rain” and bring an umbrella. For these reasons, I believe that the practical presumptive status is ‘emptied.’ Once the deliberative agent chooses the right action on epistemic grounds, the practical presumptive status does not contribute to deliberation in any substantial way. Although it may still recommend the right action, it seems redundant both from a theoretical and a practical standpoint.

187

CHAPTER 6

Table 2: Practical presumption: undercutting defeater and prior likelihoods (the Umbrella case)

1.7 The Strength of PP What determines the strength of a practical presumption? Despite many contextual, justificatory, and deontic factors, scholars typically relate a presumption’s strength to the weight of the opponent’s burden: the more demanding the opponent’s standard of proof, the stronger the presumption (Ullmann-Margalit 1983, p. 152; Rescher, 2006, p. 18; Godden & Walton, 2007, pp. 337-338). In addition to the demandingness of the opponent’s standard of proof, this dissertation identified several other factors influencing the weight of rebuttal. First, the strength of “Presumably(p), p” may vary depending on the time available to prove non-p. The opponent’s burden is heavier if he must prove non-p within minutes rather than weeks. Second, the strength of “Presumably(p), p” depends on the accessibility of evidence for non-p. Undoubtedly, the opponent’s job is more demanding if he cannot easily access evidence for non-p.180 Third, a presumption’s strength depends on the public transparency of evidence for non-p. If the evidence for non-p is novel and thereby 180

For instance, because the evidence is confidential, belongs to the distant future or past, is technical in nature, or

because getting to it requires morally problematic action.

188

Bringing It All Together: Presumption Models, and Conceptual Consequences

completely unknown to the proponent (and the general public), the opponent may carry a very weighty presentational burden. Despite having strong justification, he might need to present every piece of evidence for non-p and elaborate on it in a detailed fashion. Fourth, the strength of “Presumably(p), p” varies depending on the number of genuine attacking options the opponent has at his disposal. For instance, if the proponent infers “Presumably(p), p” from the ignorance premise “It is not proven that non-p,” then the tenability criticism of the ignorance premise (“It is untenable that ‘It is not proven that non-p’”) seems identical to proving that non-p, i.e., rebutting the conclusion “Presumably(p), p.” Therefore, in this particular case, the opponent has one genuine attacking option instead of two. All these considerations may affect the opponent’s burden and, consequently, the strength of the presumption.

2 Cognitive Presumptions (CP) 2.1 Basic Characterisation of CP Cognitive presumptions are statements with a modifier ‘presumably,’ where “Presumably(c), p” indicates p’s tentative acceptability. But although, for instance, “Andy’s cat is in the tree” (because I see it there) is defeasible, it is the most plausible truth candidate in normal circumstances (Reacher 2006). Thus, the proponent should not be obliged to defend “Andy’s cat is in the tree” in an everyday persuasion dialogue unless the opponent provides good enough argument against it. Unlike practical presumption, “Andy’s cat is in the tree” is an epistemically justified doxastic commitment that promotes the agent’s epistemic goals (acquiring empirical knowledge) for it results from a cognitive process aimed at generating true/ justified beliefs. Accordingly, a typical cognitive presumption should not be (primarily) understood as an action-oriented commitment, although a cognitive agent should proceed on “Presumably(c), Andy’s cat is in the tree” in the sense of inferring further conclusions (e.g., “Andy’s cat is not in the bedroom”). In the table below, I describe four aspects of the modifier “presumably(c).” “Presumably(c)” and “presumably(p)” differ in several respects (compare with Table 1).

6

189

CHAPTER 6

Table 3: Cognitive presumption: four aspects of the modifier ‘presumably(c)’

2.2 The Logical Function of CP Like practical presumptions, cognitive presumptions are conclusions of presumptive reasoning. Below, I provide a diagram of the core structure of presumptive cognitive reasoning by using the Cat case.

190

Bringing It All Together: Presumption Models, and Conceptual Consequences

Diagram 3: Presumptive cognitive reasoning: the core structure (the Cat case)

2.3 Justification and the Pragmatic Function of CP Unlike practical presumptions, cognitive presumptions are typically not conclusions of arguments from ignorance. Andy infers “Andy’s cat is in the tree” in the absence of a rebutting defeater, but does not derive the conclusion directly from the absence of a rebutting defeater (i.e., from the basic fact “It is not proven that Andy’s cat is not in the tree.”) The next question is: What justifies a presumptive rule? Why should Andy trust his perception and accept “Andy’s cat is in the tree?” In this dissertation, I reconstructed an economic, cost-benefit justification of presumptive rules in line with Rescher’s account. According to Rescher, when Andy faces a radical sceptic who rejects perception due to its fallibility, he can suspend judgment on all perception-based propositions. In so doing, Andy avoids the risk of adopting false perception-based beliefs. By contrast, he might trust visual perception as long as epistemic defeaters are insufficient or unknown. In so doing, Andy accepts the risk of acquiring false beliefs but also creates conditions for acquiring true perception-based beliefs. Faced with this choice, a reasonable cognitive agent should follow the policy of avoiding greater harm: since the first, sceptical solution is exceptionally costly, Andy should choose the second, more profitable solution.181

6

Practical and cognitive presumptions have different pragmatic justifications. First, deliberating and cognitive agents face distinct dilemmas since Anne, unlike Andy, cannot act on suspended judgment (taking one half of an umbrella is not a reasonable option). Second, agents are under different kinds of pressure: in the Umbrella case, pressure is 181

Despite its pragmatic flavour, Rescher’s justification of cognitive rules, for the most part, comes down to the

reliability of epistemic sources. In normal circumstances, cognitive agents should trust standard epistemic sources (perception, memory, testimony, expert opinion, etc.) because these usually generate true beliefs. In other words, if visual perception mostly generated false beliefs, then trusting visual perception would hardly be profitable from the standpoint of cognitive economy.

191

CHAPTER 6

created by external time constraints (Anne is pressured by a deadline); in the Cat case, pressure is created by internal cognitive constraints (perception is an unavoidable source of factual knowledge). Third, agents deal with distinct kinds of uncertainty; Andy’s cognitive presumption, unlike Anne’s practical presumption, is uncertain only in the sense of being defeasible, but is, otherwise, epistemically plausible. Finally, Anne and Andy rely on dissimilar considerations when comparing the costs of their actions. Whereas Anne seeks to assign unequal relevance to goals of unequal importance (health vs. convenience), Andy seeks to give equal relevance to goals of equal importance (acquiring true beliefs vs. avoiding false beliefs). As a result, practical and cognitive presumptions have different pragmatic functions. The pragmatic function of cognitive presumptions is to avoid greater harm in the face of cognitive pressure and academic (non-genuine) uncertainty. Below, I provide a diagram of the complete structure of presumptive cognitive reasoning by using the Cat case (Diagram 4).

192

Diagram 4: Presumptive cognitive reasoning: the complete structure (the Cat case)

Bringing It All Together: Presumption Models, and Conceptual Consequences

6

193

CHAPTER 6

2.4 Patterns of Interaction and the Deontic Function of CP How does a cognitive presumption influence argumentative interaction? Is the opponent immediately obliged to defend his rejection? Is the presumption’s proponent exempted from taking up the burden of proof? Cognitive presumptions impose the burden of reasoning on the opponent: after rejecting “Presumably(c), p” at t2, the opponent must either explain or argumentatively defend his view at t3. Accordingly, flat-Earther Steve may explain his denial of Diane’s standpoint “Presumably(c), the Earth is globe-shaped” by saying: “Our government fabricated the evidence that the Earth is globe-shaped.” Although Steve’s explanation is dialectically inadequate (Diane does not accept it), it is still a constructive contribution under the circumstances. For instance, after realising that, on Steve’s view, a conspiracy theory suffices to dismiss her presumption, Diane may construct a similar conspiracy theory to (adequately) challenge Steve’s standpoint: “The Flat Earth movement had the best media coverage on the day when our government passed the controversial law on public health. Don’t you find this suspicious? Our government could have fabricated the Flat Earth theory just to distract the public from the important stuff.” To be sure, using one conspiracy theory to reject the conclusion of another, equally bizarre conspiracy theory is not a hallmark of epistemic rationality. But in the context of persuasion dialogue, Diane’s hands are tied. Once Steve rejects “Presumably(c), the Earth is globe-shaped,” she does not have many epistemically plausible reasons available (such as scientific evidence) that would persuade a flat-Earther. Moreover, since flat-Earthers refuse well-known evidence, Diane can do nothing but guess what reasons might convince Steve. In such circumstances, Steve’s explanation gives Diane a fighting chance: it reveals what considerations she needs to attack and how to develop a dialectically adequate (but not epistemically ideal) strategy. In the profile below, I illustrate Steve’s burden of reasoning.

194

Bringing It All Together: Presumption Models, and Conceptual Consequences

Figure 3: Cognitive presumption: a profile of a dialogue (the opponent Steve / the Flat Earth example)

If Steve decides to argue at t3, he must produce a relevant (and adequate) reason, thereby taking up the burden of production. By satisfying this burden, Steve does what is necessary to ‘stay in the game’ and prevail in the discussion at some later point. Importantly, at t3, he is allowed to offer some minor argument, to get things going (e.g., “According to modern science, the Earth is globe-shaped, but science is often wrong”).

6

Ultimately, however, Steve carries the burden of successful defence. Just like the burden of persuasion in the practical case, the burden of successful defence is allocated in the opening stage, influences the interaction in the argumentation stage, and is evaluated in the concluding stage. It entails that Steve must have sufficient argument to win the discussion in the concluding stage. Crucially, unlike the burden of persuasion, the burden of successful defence does not entail that those who shoulder it lose debates if, ultimately, their arguments are insufficient. Suppose that Steve’s arguments for “The Earth is disc-shaped” are as strong as Diane’s arguments for “The Earth is globe-shaped” in the concluding stage. If the goal of the dialogue is to accept a more convincing view, the uncertainty should not count in Diane’s favour. Instead, from a dialectical standpoint, parties should call it a draw and suspend judgment. If there is no clear winner, it remains an open question whether extra-dialectical considerations—like the objective truth that “The Earth is globe-shaped,” its God’s eye justification, its acceptance among Steve’s epistemic peers, the importance of intellectual humility, etc.—should still commit Steve to accept Diane’s viewpoint.

195

CHAPTER 6

Next, whether the presumption’s proponent takes up the burden of proof depends on the transparency of her justification. If Diane and Steve have previously accepted “Presumably, the Earth is globe-shaped” based on scientific evidence, and Steve retracts his commitment at t2, then Diane is not obliged to provide the apparent justification at t3. Of course, if Steve were to present a strong argument against “Presumably, the Earth is globe-shaped,” then Diane would be imposed with the burden of production. Technically, however, her probative obligation would not concern cognitive presumption. From a purely dialectical perspective, given Steve’s strong argument at t3, “The Earth is globeshaped” would not be the most plausible proposition at t4. Hence, if the justification is transparent and the opponent’s argument is weak, the proponent is not obliged to defend “Presumably(c), p” and cognitive presumption entails binary asymmetry. However, the proponent becomes obliged to defend p, which no longer has the status of a presumption, by virtue of a strong argument offered by the opponent at t3. This is yet another difference between practical and cognitive presumptions. In the profile below, I illustrate the interaction pattern where the presumption’s proponent is imposed with the burden of production.

Figure 4: Cognitive presumption: a profile of a dialogue (the proponent Diane / the Flat Earth example)

196

Bringing It All Together: Presumption Models, and Conceptual Consequences

As a result, we can formulate the so-called deontic function of cognitive presumption as follows: Production of arguments (in the argumentation stage) a) The proponent does not shoulder the burden of production for “Presumably, p.” o If reasons for “Presumably, p” are entirely transparent. b) The opponent rejecting “Presumably, p” is obliged to provide either an explanation or an argument. Initially, at t3, both moves are (equally) constructive. Evaluation of arguments (in the concluding stage) a) The proponent does not shoulder the burden of successful defence for “Presumably, p” (as long as p is the most plausible proposition, and its justification is entirely transparent.) b) The opponent rejecting “Presumably, p” shoulders the burden of successful defence. Cognitive presumptions are also associated with degree asymmetries concerning evidence accessibility and commitment strictness. Since the evidence for cognitive presumptions (e.g., “The Earth is globe-shaped,” “The cat is in the tree,” or “Smoking is unhealthy”) is usually easily accessible, the proponent carries a lighter burden than the presumption’s opponent. It is a well-known fact that, nowadays, it is easier to find and present good reasons for “The Earth is globe-shaped” than for “The Earth is disc-shaped.” In addition, due to the typical accessibility of justifications for presumptions, the proponent may not be obliged to present these justifications. Unlike the presumption’s opponent, the proponent may have a less strict commitment and be recommended (instead of obliged) to give reasons.

6

2.5 Dialogical Functions of CP How do cognitive presumptions benefit the dialectical procedure? I distinguish several dialogical functions of cognitive presumptions. First, since cognitive presumptions are generated by reliable epistemic sources and are usually employed in epistemic dialogue (inquiry), they contribute to acquiring empirical knowledge and justified belief. I label this prominent dialogical role as cognitive presumption’s ultimate contextual function.182 Second, since most cognitive presumptions are uncontroversial and well-known (“The Earth is globe-shaped,” “Smoking is unhealthy,” “Cereal is not soup,” etc.), they usually represent dialogical starting points. Dialogical parties must share common premises (identified in the opening stage) to engage in reasonable persuasion, so cognitive presumptions constitute foundations of argumentative dialogue and allow the discussion to take off. 182  Obviously, cognitive presumptions can also contribute to non-epistemic dialogues. For instance, if Anne knows that “It will rain” represents the most plausible proposition, she will take an umbrella due to a cognitive presumption (and not due to a practical one).

197

CHAPTER 6

Third, even if p is not a shared premise in a dialogue, its presumptive status may contribute to a fruitful discussion. Since the justification of some widely-shared proposition p is usually transparent, presumption saves time by exempting the proponent from the burden of proof—the proponent is not obliged to ‘cover known ground.’ Additionally, by requesting reasons from an eccentric opponent, the presumption’s deontic function helps the proponent to become familiar with the unorthodox view and gives her a chance to develop a dialectically adequate attack strategy. Fourth, if p is not a shared premise and is challenged by an eccentric opponent, its presumptive status may prevent dialectical regress. Namely, after Diane puts forward “The Earth is globe-shaped,” Steve may try to win the discussion by requesting justification for every new reason introduced by Diane. By imposing the burden on Steve, the presumptive status of “The Earth is globe-shaped” sets limitations to this annoying strategy. The roles of furnishing starting points, enhancing the discussion’s effectiveness, and preventing dialectical regress, I label the ‘dialogical functions’ of cognitive presumption. Moreover, cognitive presumptions may have a pre-dialogical function because they specify conditions under which their advocates should consider having a debate, to begin with. Namely, why would Diane discuss the Earth’s shape at all? Since “The Earth is discshaped” is highly implausible, she cannot honestly expect that exploring the flat-Earth hypothesis will improve her epistemic position. From an epistemic standpoint, exploring this hypothesis seems like a waste of time. Nevertheless, Steve may motivate Diane to consider his controversial standpoint by immediately supporting the flat-Earth hypothesis. By giving reasons, Steve shows Diane that he is a serious interlocutor and that discussing the flat-Earth theory may not be a waste of time. This provides another explanation for why the opponent takes up the burden of reasoning. In the debate, Steve must present reasons to persuade Diane to accept his view. But before the debate, he must provide reasons to motivate Diane to consider discussing his position in the first place (for similar points, see Kauffeld 1998; Aijaz et al 2013).

2.6 Conditions of CP’s Defeat If cognitive presumptions stand until proven otherwise, how can the opponent discharge the burden of proof? In the Cat case, how can Jane defeat “Presumably(c), Andy’s cat is in the tree?” Just like in the case of practical presumption, the opponent may rebut cognitive presumption p, engage in various kinds of premise tenability criticism, or focus on some connection criticism (undermining, undercutting, etc.). However, this dissertation explored the circumstances in which a particular type of connection criticism—the undercutting defeater—cancels a cognitive presumption. Suppose that Jane says: “The neighbour has a cat that looks exactly like Andy’s” (u). Her evidence undercuts the connection between the basic fact “Andy sees that a cat, which looks exactly like his cat, is in the tree” (q) and the conclusion “Andy’s cat is in the tree” (p). However, u’s potential to defeat “Presumably(c), p” depends on p’s prior likelihood. 198

Bringing It All Together: Presumption Models, and Conceptual Consequences

First, if Andy and Jane attribute equal likelihoods to “Andy’s cat is in the tree” and “Andy’s cat is not in the tree,” and Jane introduces the aforementioned undercutter, Andy’s conclusion is not the most plausible proposition anymore: its presumptive status is defeated. Second, if the prior likelihood of “Andy’s cat is in the tree” is sufficiently higher than the prior likelihood of “Andy’s cat is not in the tree,” Andy’s conclusion remains the most plausible, and its presumptive status is not defeated. Third, if the prior likelihood of “Andy’s cat is in the tree” is sufficiently lower than the prior likelihood of “Andy’s cat is not in the tree,” Jane successfully undercuts the conclusion’s presumptive status. Only in the third scenario do cognitive presumptions “react” to undercutting defeaters in the same way practical presumptions do. Table 4 summarises the relation between cognitive presumptions, undercutters, and prior likelihoods.

6

Table 4: Cognitive presumption: undercutting defeater and prior likelihoods (the Cat case)

199

CHAPTER 6

2.7 The Strength of CP From a purely epistemic standpoint, the strength of “Presumably(c), p” depends on p’s justification and is correlated to p’s plausibility. However, in a dialectical setting, various factors influencing the opponent’s burden affect a presumption’s strength. First, the strength of “Presumably(c), p” depends on the time available to prove non-p. Second, it depends on the accessibility of evidence for non-p. Third, it is influenced by the public transparency of evidence for non-p. Fourth, the strength of “Presumably(c), p” varies depending on how many genuine attacking options the opponent has at his disposal. Nevertheless, this dissertation showed that the strength of cognitive presumption, unlike the strength of practical presumption, is not entirely correlated to the weight of rebuttal. Sometimes, the fact that cognitive presumption p became weaker (less plausible) does not entail that the opponent’s burden of proving non-p became less demanding.

3 What Should the Concept of Presumption Amount to? So, practical and cognitive presumptions differ in many respects. To make the differences salient, let’s compare their features in Table 5.

200

Bringing It All Together: Presumption Models, and Conceptual Consequences

6

Table 5: Practical and cognitive presumptions: summary and comparison

What do these differences imply? On standard accounts, do practical and cognitive presumptions still share common features? Is identifying shared features theoretically helpful? And does the commitment associated with cognitive presumption, given all these differences, fit the label of ‘presumption?’ As explained in introductory chapter, the word ‘presumption’ can have several meanings in everyday and legal contexts. Let’s focus on the following ones: 1. A presumption denotes having “a strong reason for believing something to be so” (Merriam-Webster.com Dictionary). 2. A presumption is an “act of believing that something is true without having any proof” (Cambridge Advanced Learner’s Dictionary). 3. A presumption is a “conclusion derived from a particular set of facts based on law, rather than probable reasoning” (The American Heritage Dictionary of the English Language). 201

CHAPTER 6

Standard theoretical accounts of presumption perpetuate the ambiguity associated with everyday and legal meanings.183 While the characterisation of cognitive presumption corresponds with the first meaning, the characterisation of practical presumption, for the most part, corresponds with the second and third meaning. Since the theoretical concept of presumption is fragmented, much like its everyday and legal counterparts, standard accounts fail to (semantically) improve their central notion. Inheriting some of this pre-existing ambiguity in philosophy and argumentation theory is unfortunate, but whether it is truly problematic depends on the theorist’s ambition. If the argumentation theorist seeks to develop a logical and dialogical model of presumption, the conceptual incoherency is harmless. Namely, after recognising an expression’s ambiguity, the theorist should acknowledge it and treat practical and cognitive presumption separately. Moreover, for a philosopher who conceptually analyses and explores laymen’s intuitions about common expressions, the incoherency of presumptions is commonplace. After all, many linguistic expressions have multiple meanings (e.g., knowledge that vs. knowledge how; intuition as a mere hunch vs. intuition as an obvious, true proposition, etc.), but homonymy is harmless as long as we can identify the intended meaning in the context at hand. The philosopher, then, should analyse some notion’s specific meaning (What is ‘knowledge that?’, What is ‘practical presumption?’, etc.) and hope for coherency within this limited scope.

3.1 The Limitations of Normative and Functional Conceptions of Presumption However, instead of building on pre-existing conceptual and linguistic intuitions, an ambitious theorist may seek to develop a theoretically optimal language in which distinct expressions denote distinct phenomena. In carrying out this disambiguation the theorist might first decide to pin the term (concept of) ‘presumption’ onto only one relevant kind of proposition or commitment. Second, she might choose to abandon the word (concept of) ‘presumption’ altogether. Inspired by Ullmann-Margalit (1983), Godden (2019) defends the first solution and argues that ‘presumably’ should denote only practical presumptions. In his view, when understood as a purely procedural, action-oriented qualifier, ‘presumably’ expresses a unique commitment—associated with a particular case of instrumental reasoning —that cannot be represented by other qualifiers. By contrast, if associated with cognitive presumptions, ‘presumably’ indicates a provisional epistemic commitment that can be expressed by other qualifiers, such as ‘plausibly’ or ‘defeasibly.’ Similarly, the legal scholars advocating the “normative conception” of presumption maintain that “the term ‘presumption of fact’ should be discarded as useless and confusing”184 (Wigmore, as cited 183 

For a detailed analysis of presumption’s ambiguity in the legal discourse, see Gama (2017, pp. 562-569).

184  In many relevant respects, the legal concept of ‘presumption of fact’ corresponds with our concept of ‘cognitive presumption.’

202

Bringing It All Together: Presumption Models, and Conceptual Consequences

in Gama 2017, p. 565). Instead, we should set the word ‘presumption’ aside for inferences supported by legal rules. The more extreme solution is to completely abandon the expression ‘presumption,’ since, on the one hand, it creates confusion and, on the other, does not contribute to resolving theoretical problems. According to Gama, Ronald Allen—the leading proponent of the “functional conception” of presumption in legal scholarship—argues that “the word ‘presumption’ could be completely eliminated,” since “the term ‘presumption’ is applied without adding any significance to the solution of the problem” (Gama 2017, p. 567). Allen (1980, 2014) believes that legal scholars should resolve evidential problems associated with incompatible meanings of ‘presumption’ and stop looking for essential features that unify all presumptions. In argumentation theory, Lewiński (2017) makes a similar point. He argues that “the study of presumption always collapses into the study of other, likely more fundamental, concepts” (p. 610) because of presumption’s “epiphenomenal character” (p. 596). So, while presumption—both as a term and as a concept—generates some problems, it is unclear whether presumption provides any solutions. To my mind, both responses to the incoherency of presumptions are too extreme. Allen is right that scholars should not get too distracted by linguistic and conceptual issues, but ‘presumption’ is not an empty and useless label. As Godden (2017, 2019) observes, the practical version of ‘presumably’ indicates a special commitment and appears irreducible to other qualifiers. Since ‘presumably’ is a singular expression that, in the special case of instrumental reasoning, explains the unique (procedural) nature of an agent’s commitment, it seems reasonable to preserve it in theoretical language. Practical presumption remains a genuine challenge for any reductionist about presumptions.

6

However, Godden’s thesis that the cognitive version of ‘presumably’ is redundant, since it can be reduced to ‘plausibly’ or ‘defeasibly,’ appears somewhat premature. Recall that cognitive presumptions are not just plausible propositions but the most plausible propositions (Rescher 2006). In dialogue, cognitive presumption is not just a defeasible starting point, but the most plausible defeasible starting point. This singular epistemic status justifies whatever function cognitive presumption is supposed to perform. For instance, cognitive presumption p would neither shift the burden of proof nor terminate dialectical regress if p were accompanied by equally plausible but contradictory proposition q. Thus, since the epistemic modifier ‘presumably’ is more specific than qualifiers ‘plausibly’ and ‘defeasibly,’ it might be reasonable to keep it in theoretical language. The theorist may avoid the potential confusion of using the same expression to denote distinct kinds of commitments most explicitly, e.g., by distinguishing ‘presumably(p)’ from ‘presumably(c).’

203

CHAPTER 6

3.2 Towards a Pluralistic Conception of Presumption When the meaning of ‘presumption’ is clear in a given context, discussing whether to prefer ‘presumption’ over some other word seems quite inconsequential. But when the conceptual (rather than the linguistic) aspect of the debate is concerned, I believe that standard accounts should adopt a “pluralistic conception” of presumption: [T]he proponents of this conception [in the legal scholarship] argue that the appropriate way to analyze the concept of presumption is through the examination of the paradigmatic types of presumption used in legal language. As a result of this operation, there would be not a single concept of presumption, but a variety of concepts of presumptions. Thus, the question “What is a presumption?” does not admit a single response in this conception, but as many as suggested by the specific instances in which the word “presumption” is applied. (Gama 2017, p. 566) Although the concepts differ and Rescher is careless in portraying cognitive presumption as “the epistemic analogue of ‘innocent until proven guilty’” (2006, p. 23), there may still be hope for identifying general features common to all presumptions. The theorist might consider at least four shared features: 1. Practical and cognitive presumptions are actionable commitments, since deliberating/ the cognitive agent is recommended—on either epistemic or non-epistemic grounds— to proceed as if the presumption were true. It is debatable whether assumptions (stipulations) or hypotheses are actionable in a comparable sense. 2. Presumptions can be reconstructed as conclusions of reasoning, unlike, for instance, assumptions. 3. On a pragmatic interpretation, presumptions are always associated with uncertainty, pressure, and harm reduction. This is not always true for assertions, assumptions, and hypotheses. 4. Finally, when challenged, both types of presumptions impose the burden of reasoning on the opponent. Assertions, assumptions, and hypotheses entail different distributions of dialectical obligations. As a result, I believe that standard accounts should formulate the concept of presumption in terms of a moderate pluralism. They should treat practical and cognitive presumptions as different concepts as long as they investigate the applications of presumption in the dialogical contexts: the two types of presumption represent different kinds of modifiers, with distinct pragmatic, deontic, and dialogical functions, dissimilar defeating conditions, and different conceptions of strength. Nevertheless, practical and cognitive presumptions become more similar and may even merit a joint treatment once we compare them to other pragmatic phenomena.

204

Bringing It All Together: Presumption Models, and Conceptual Consequences

4 Conclusion Practical and cognitive presumptions entail distinct contextual, pragmatic, dialogical, and deontic functions; different defeating conditions; and distinct conceptions of ‘strength.’ Thus, one should analyse them separately as long as one investigates applications in dialogical contexts. However, the concept of ‘presumption’—taken in the wide sense— may still serve some theoretical purpose since practical and cognitive presumptions look quite similar to each other when compared to other propositional attitudes (such as assertions, assumptions, or hypotheses). Accordingly, I advocate a ‘moderate pluralism’ about presumptions.

6

205

C

CONCLUSION

CONCLU SI ON

There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy. —Shakespeare, Hamlet

208

Conclusion This dissertation contributed to traditional, legally-inspired studies of presumption in philosophy and argumentation theory. Its first, ‘dialectical objective’ was to review, reconstruct, evaluate, and improve traditional models of presumption. Its second, ‘typological objective’ was to compare practical and cognitive presumptions in systematic fashion in order to gain a deeper understanding of their relationships. The third, ‘conceptual objective’ was to clarify the essential concepts of presumption studies (‘uncertainty,’ ‘ignorance,’ ‘pressure,’ ‘harm,’ ‘burden of proof,’ ‘asymmetry,’ ‘strength,’ etc.) and, consequently, elucidate the concept of ‘presumption’ itself. To achieve these theoretical objectives, I explored seven research questions. 1. What is a presumption, ontologically speaking? Is it a proposition, a status of a proposition, a relation, or a speech act? 2. What is the logical function of a presumption? Is it a rule or a conclusion of presumptive reasoning? 3. How is a presumption justified? Can one reduce presumptive reasoning to an argument from ignorance? Do presumptions avoid greater harm, and how does this pragmatic function connect to underlying values? 4. What is the deontic function of a presumption? Must challengers of presumptions offer arguments, or can they (also) offer explanations? Are the proponents of presumptions always exempted from taking up the burden of proof? 5. What are the dialogical functions of presumption? How exactly does dialectical procedure benefit from asymmetrical allocations of the burden of proof? 6. What are the defeating conditions of presumptions? What types of evidence defeat presumptions, and in which evidential circumstances can a particular defeater efficiently cancel a presumption?

C

7. What contextual, justificatory, and deontic factors make presumption p stronger or weaker? And is the weight of the Challenger’s burden of rebuttal—i.e., how difficult it is for the Challenger to prove non-p—always related to the strength of p’s justification?

209

CONCLU SI ON

1 A Model of Practical Presumption In developing a systematic functional account of practical presumption, I primarily relied on the works of Ullmann-Margalit (1983), Godden (2017, 2019), and Walton (2014). Practical presumption p conveys the procedural and non-epistemic status of proposition p: if p is a practical presumption, a deliberating agent should proceed as if p were true due to non-epistemic reasons. For instance, if chances of rain are uncertain and Anne must leave her apartment, she should aim to avoid greater harm. Since getting soaked represents greater harm compared to needlessly carrying an umbrella, Anne proceeds as if it will rain, and brings an umbrella, i.e., she concludes “Presumably(p), it will rain.” Since practical presumptions avoid greater (non-epistemic) harm in circumstances of (deliberative) pressure and (genuine) uncertainty, we can reconstruct “Presumably(p), p” as the conclusion of presumptive practical reasoning. Notably, presumptive practical reasoning closely resembles the so-called ‘argument from ignorance.’ Upon rejecting a practical presumption in a dialogue’s argumentation stage, the Challenger incurs the ‘burden of reasoning:’ she must present either an argument or an explanation in support of her rejection. Due to practical stakes, uncertainty, and time pressure, however, the Challenger is strongly recommended to assume the ‘burden of production’ and present an argument. Paradigmatically, arguments are tailored for persuasion. Unlike typical explanations, they can change the proponent’s opinion about the relevant issue, resolve dialectical uncertainty, and shift the default course of action. Crucially, the Challenger also carries the ‘burden of persuasion’ in the concluding stage: for instance, when the time of decision comes, if the Challenger cannot prove “It will not rain,” he loses and Anne wins. As a result, agents should conclude “Presumably(p), it will rain” and bring an umbrella. The ultimate contextual function of practical presumptions is choosing the right action. But their special dialogical function is facilitating dialogical progress by enabling resolution in the concluding stage. Should evidence that normally governs our actions remain uncertain when the time of decision comes, proceeding as if the presumed claim were true avoids both deliberative paralysis, as well as random decision-making (e.g., choosing the action by tossing a coin). Next, while the Challenger can cancel “Presumably(p), p” in many standard ways— e.g., by proving non-p, or criticising the tenability of some premise of presumptive reasoning—usually, she cannot achieve this by using an ‘undercutting defeater.’ An undercutting defeater attacks the connection between premise q and conclusion p and, typically, shows that a reasonable epistemic agent should suspend her belief in conclusion p. However, adopting an agnostic stance towards p seems fully compatible with p being a practical presumption. In fact, in paradigmatic cases, p becomes a practical presumption exactly at times when one does not know whether p is true or not. In the Umbrella case, Anne treats “It will rain” as a practical presumption precisely because rain is uncertain.

210

An undercutter could defeat a practical presumption, but only on rare occasions, and given very specific background information. Finally, a practical presumption is stronger or weaker depending on various contextual, justificatory, and deontic factors. On standard views, contextual and justificatory factors determine deontic considerations: the higher the stakes and values justifying “Presumably(p), p,” the stronger the presumption p, and the more difficult it is for the Challenger to prove non-p. For instance, since “Presumably(p), wearing a face mask protects other people from contracting COVID-19” reduces the risk of dangerous infections, and “Presumably(p), it will rain” avoids the risk of getting soaked, the former presumption is stronger. As a result, while the Challenger must present substantial evidence to rebut the first presumption, he can rebut the second presumption by presenting a relatively weak argument. This dissertation contributed to studying practical presumptions by (1) developing a detailed scheme of presumptive practical reasoning; (2) stressing that, immediately after rejecting “Presumably(p), p” at t2, the Challenger is strongly recommended (but not obliged!) to provide an argument at t3; (3) explaining why the Challenger usually cannot overthrow “Presumably(p), p” by using an undercutting defeater; and, finally, (4) developing a systematic overview of factors that affect the strength of “Presumably(p), p.”

2 A Model of Cognitive Presumption In developing a systematic functional account of cognitive (epistemic, initial) presumption, I primarily relied on the works of Rescher (1977, 2006), Freeman (2005), and Walton (2014). Cognitive presumption p primarily conveys the doxastic and epistemic status of p: if p is a cognitive presumption, an agent should take p as true (and proceed accordingly), for p is generated by some reliable epistemic source (e.g., perception, memory, testimony, expert opinion, etc.), and, in the context at hand, represents the most plausible proposition.

C

To some degree, taking p as true appears consistent with avoiding greater harm in epistemic inquiry. Suppose Andy sees that “Andy’s cat is in the tree.” Being aware of perception’s fallibility, he is under pressure to decide whether to (1) trust his senses, conclude “Presumably(c), Andy’s cat is in the tree,” and derive further conclusions from it, or (2) adopt a sceptical stance and suspend judgment. In the long run, trusting reliable sources—despite the cost of adopting some false beliefs—appears less harmful than repeatedly suspending judgment. From a standpoint of cognitive pragmatism, cognitive presumptions avoid greater (epistemic) harm in circumstances of (cognitive) pressure and (academic/non-genuine) uncertainty. Thus, we can reconstruct “Presumably(c), p” as the conclusion of presumptive cognitive reasoning. Notably, presumptive cognitive reasoning does not strongly resemble the so-called ‘argument from ignorance.’ Upon rejecting a cognitive presumption at t2, the Challenger incurs the ‘burden of reasoning’ at t3. Crucially, at t3, presenting an explanation may be equally legitimate 211

CONCLU SI ON

as presenting an argument. Imagine that flat-Earther Steve rejects Diane’s standpoint “Presumably(c), the Earth is globe-shaped.” Since flat-Earthers refuse well-known (scientific) evidence, Diane can do nothing but guess what reasons might convince Steve. In such circumstances, Steve’s explanation may serve as a strategic guideline that will help Diane to develop a dialectically adequate strategy, and, thus, be a constructive contribution. Also, the Challenger carries the ‘burden of successful defence’ in the concluding stage. In other words, Steve must provide sufficient argument at some point to win the discussion. Crucially, the burden of successful defence (associated with cognitive presumption) differs from the burden of persuasion (associated with practical presumption) because challengers who shoulder it do not automatically lose debates if their arguments are insufficient. If, ultimately, Steve’s arguments for “The Earth is disc-shaped” and Diane’s arguments for “Presumably(c), the Earth is globe-shaped” are equally plausible, the uncertainty should not count in Diane’s favour. The ultimate contextual function of cognitive presumptions is attaining empirical knowledge. Their special dialogical functions include facilitating dialogical progress (by furnishing starting points in the opening stage), preventing dialectical regress (by setting limitations to the persistent interlocutor’s strategy), and enhancing the dialogue’s fruitfulness (by avoiding repetition of already known arguments). Next, the Challenger can cancel “Presumably(c), p” in many standard ways, including by use of an ‘undercutting defeater.’ For instance, if Andy concludes “Presumably(c), Andy’s cat is in the tree” because “Andy sees that a cat, which looks exactly like his cat, is in the tree,” Jane can defeat his presumptive reasoning by saying: “Andy, the neighbour has a cat that looks exactly like yours.” All else being equal, this information makes “Andy’s cat is in the tree” and “The neighbour’s cat is in the tree” equally plausible, and since, by definition, cognitive presumption is the most plausible proposition, its original presumptive status is lost. Therefore, unlike practical presumptions, undercutting defeaters usually cancel cognitive ones. Finally, a cognitive presumption is stronger or weaker depending on many contextual, justificatory, and deontic factors. But contextual and justificatory factors do not always determine deontic considerations. The fact that cognitive presumption p became weaker (less plausible) does not entail that the Challenger’s burden of proving non-p became less demanding. This reveals a possible structural difference in explaining the strength of practical and cognitive presumptions. This dissertation contributed to studying cognitive presumptions by (1) distinguishing cognitive presumptions from arguments form ignorance; (2) developing a detailed scheme of presumptive cognitive reasoning; (3) showing that the policy of avoiding greater harm, when associated with cognitive presumptions, rests on structurally different considerations than when associated with practical presumptions; (4) stressing that, after rejecting “Presumably(c), p” at t2, the Challenger is obliged to provide an argument, or an explanation at t3, and that both moves are legitimate; (5) clarifying that the Challenger of 212

a cognitive presumption carries the ‘burden of successful defence’ in the concluding stage (instead of the ‘burden of persuasion’); and, finally, (6) systematising factors affecting the strength of “Presumably(c), p,” and showing that, sometimes, the presumption’s strength is not correlated with the weight of the Challenger’s burden of proof.

3 The Concept of Presumption I argued that the differences between practical and cognitive presumptions merit serious recognition. As long as we investigate the applications of presumption in dialogical contexts, we should treat practical and cognitive presumptions as separate concepts. However, in my view, the concept of ‘presumption’ is still informative enough to serve a theoretical purpose. That is, although practical and cognitive presumptions seem quite different when compared to each other, they may look quite similar when compared to other propositional attitudes or speech acts (such as assertions, assumptions, or hypotheses). They are both actionable commitments; can be reconstructed as conclusions of reasoning; are associated with uncertainty, pressure, and harm reduction (on a pragmatic interpretation, at least); and, when challenged, both impose the burden of reasoning on the Challenger. Accordingly, I argue for—what might be called—a ‘moderate pluralism’ about presumptions.

4 Guidelines for Further Research This dissertation sought to develop a systematic theory of presumption by critically exploring the standard accounts. Unsurprisingly, the research is still ongoing. First, future research should give more detailed answers to some questions that are directly associated with the proposed framework. 1. What exactly is the relationship between practical presumption and epistemic uncertainty? Must p, by definition, be uncertain to have the status of practical presumption, or is uncertainty just a typical feature?

C

2. Is pragmatism necessary to explain the justification of cognitive presumption? Is trusting standard epistemic sources reasonable, for being a suitable means to an epistemic end or because standard sources are reliable? And does pragmatic justification (expressed in the language of epistemic policies) collapse into a statistical one? For instance, is trusting testimonies profitable only if the testimony is truth-conducive? Would trusting testimony be epistemically profitable in any way if most testimonies were false? 3. What does (pragmatic) justification of cognitive presumption look like when the opponent is not a radical sceptic (a person that advocates the suspension of judgment) but the second proponent who advances a contradictory view? Does this modify the presumptive rule’s justification, as presented in our complete scheme of cognitive reasoning?

213

CONCLU SI ON

4. How do we justify a practical presumption when non-epistemic values of equal importance govern deliberation? And can some epistemic value of higher priority justify a cognitive presumption? For instance, can “The Earth is globe-shaped” be presumed because, in some context, intellectual humility (“People who know better than me think so”) is treated as a higher epistemic value than intellectual autonomy (“Well, the Earth is flat, and I have my reasons…”)? Also, future research should develop a more sophisticated metatheory about the connection between non-dialectical and dialectical normativity. 5. On what grounds, exactly, does non-dialectical normativity become dialectically relevant? In objective terms, “The Earth is globe-shaped” is epistemically justified. But why should its objective justification be dialectically pertinent and influence the flat-Earther’s obligations and entitlements if the flat-Earther does not recognise it? 6. In what way, exactly, is non-dialectical normativity dialectically relevant? Is the objective justification of “The Earth is globe-shaped” supposed to determine the initial structure of dialogue in the following sense: while the flat-Earther must produce reasons, the round-Earther is allowed only to criticise them? Or is it supposed to determine the final result of the discussion in the following sense: if the flat-Earther fails to prove “The Earth is disc-shaped,” he must accept “The Earth is globe-shaped?” This dissertation offered some preliminary answers, but they require additional justification. Second, future research should tackle questions that follow up on the current project. 1. How do the proposed schemes of presumptive practical and cognitive reasoning connect to Walton’s theory of argumentation schemes and critical questions? For instance, I did not explore the idea that the opponent’s burden of proof could depend on the particular critical question the opponent asks while challenging some argumentation scheme. However, different critical questions might allocate the burden of proof in different ways (see Walton, Reed & Macagno 2008, p. 31; Atkinson et al. 2020, p. 288, 304, 311). 2. What is the connection between presumptions and various argumentation fallacies? Although this dissertation linked (rejecting) presumptions with evading or shifting the burden of proof, relations between presumptions and fallacies are yet to be explored. 3. Future research should analyse rhetorical dimensions of presumption. How do presumptions promote the goal of effective persuasion? Can it be strategically reasonable to accept the burden of proof even when the presumption’s proponent is not dialectically obliged to do so? And what would be the rhetorically expedient strategies to discharge such a burden (see van Eemeren & Houtlosser 2002)?

214

Future research should incorporate insights from speech-act theory to fully realise conceptual, dialectical, and typological objectives. 4. Is presumption a verdictive (Corredor 2017, p. 573), a supposition (Kauffeld 2003, p. 139), an assumption (Bermejo-Luque 2016, p. 9), an assertive (van Eemeren & Houtlosser 2002, p. 17), an assertion (Godden 2019, p. 217), or something between an assertion and an assumption (Walton 1993, p. 125)? 5. What is the contribution of a particular speech-act characterisation to explaining the dialectical norms associated with presumptions? 6. Can a speech-act characterisation highlight or deepen the differences between practical and cognitive presumptions? However, the greatest potential lies in connecting this abstract research on presumptions to real-world issues and other academic disciplines. For example, in social controversies, are presumptions only a conservative’s tool (Räikkä 1997), or are there liberal presumptions as well (Goodnight 2019)? Given the recent popularity of conspiracy theories in Western societies, can we talk about the post-presumption model of argumentation (Neville-Shepard 2019)? This dissertation sought to exploit the resources of argumentation theory and, to a modest extent, contemporary epistemology. The further step is to take a more interdisciplinary approach and incorporate insights from related disciplines.

C

215

CONCLU SI ON

216

Summary

Summary On the standard accounts, presumptions are defeasible propositions which asymmetrically allocate the burden of proof. Scholars typically distinguish ‘practical’ from ‘cognitive’ presumptions. Practical presumptions are propositions the deliberating agents should proceed on due to non-epistemic reasons. For instance, if chances of rain are uncertain, Anne should aim to avoid greater non-epistemic harm (getting soaked), proceed as if “It will rain,” and bring an umbrella. By contrast, cognitive presumptions are propositions the cognitive agents should proceed on due to epistemic reasons. For instance, if Andy sees that his cat is in the tree, he should trust his senses, conclude “Andy’s cat is in the tree,” and derive further conclusions from it. Despite the cost of adopting some false beliefs, trusting reliable sources is less harmful, i.e., epistemically profitable in a long run. This dissertation contributes to philosophy and argumentation theory by exploiting the resources of traditional accounts of presumption. It refines them by clarifying claims, reconstructing missing parts, and adding conceptual and typological detail. It focuses on the following questions: 1. How is a presumption justified? Is presumptive reasoning reducible to an argument from ignorance? Do all presumptions avoid greater harm (pragmatic function)? 2. Must challengers of presumptions offer arguments, or can they (also) offer explanations? Are the proponents of presumptions always exempted from taking up the burden of proof (deontic function)? 3. How does dialectical procedure benefit from asymmetrical allocations of the burden of proof (dialogical function)? 4. What types of defeaters cancel presumptions, and in which evidential circumstances (defeating conditions)?

C

5. What factors make presumption stronger or weaker (strength conceptions)? The dissertation hopes to achieve three general objectives: a) elucidate concepts essential to standard accounts of presumption (such as ‘argument from ignorance,’ ‘the deliberation problem,’ ‘evidential uncertainty,’ ‘burden of proof,’ ‘asymmetrical allocation,’ ‘strength,’ etc.); b) develop dialogical model of presumption (e.g., find out which patterns of interaction are associated with presumptions); and c) further explore the relationship between practical and cognitive presumptions. The dissertation shows that two types of presumption entail distinct contextual, pragmatic, dialogical, and deontic functions; different defeating conditions; and distinct conceptions of ‘strength.’ Thus, one should analyse practical and cognitive presumptions separately as long as one investigates their applications in dialogical contexts. But since practical and cognitive presumptions look similar when compared to other propositional

217

CONCLU SI ON

attitudes (such as assertions, assumptions, or hypotheses), the concept of ‘presumption’ may still serve some theoretical purpose. Accordingly, I argue in favour of a ‘moderate pluralism’ about presumptions.

218

Samenvatting

Samenvatting Volgens standard benaderingen zijn presumpties weerlegbare proposities die gepaard gaan met een asymmetrische verdeling van de bewijslast. Onderzoekers onderscheiden gewoonlijk ‘praktische’ en ‘cognitieve’ presumpties. Praktische presumpties zijn nietepistemisch gemotiveerde proposities die ten grondslag liggen aan de keuzes van een handelende persoon. Bijvoorbeeld, als het onduidelijk is of het gaat regenen moet Anne proberen het grotere risico van niet-epistemische schade (doorweekt raken) voorkomen, handelen alsof “het gaat regenen”, en een paraplu meebrengen. Cognitieve presumpties daarentegen zijn epistemisch gemotiveerde proposities die ten grondslag liggen aan de keuzes van een denkende persoon. Als Andy bijvoorbeeld ziet dat zijn kat in de boom zit moet hij vertrouwen op zijn zintuigen, concluderen dat “Andy’s kat in de boom zit,” en er verdere conclusies aan verbinden. Hoewel het een onware gevolgtrekking kan opleveren is het vertrouwen op deugdelijke bronnen minder nadelig, dat wil zeggen, epistemisch lonend op de lange termijn. Dit proefschrift draagt bij aan de filosofie en de argumentatietheorie door gebruik te maken van het potentieel van traditionele besprekingen van presumptie. Het verfijnt ze door standpunten op te helderen, ontbrekende onderdelen te reconstrueren, en door conceptuele en typologische details aan te brengen. Het richt zich met name op de volgende vragen: 1. How is een presumptie gerechtvaardigd? Is presumptief redeneren herleidbaar tot een beroep op onwetendheid (argumentum ad ignorantiam)? Vermijden alle presumpties grotere schade (de pragmatische functie van presumptie)? 2. Moeten degenen die presumpties kritisch bevragen argumentatie aandragen, of mogen ze (ook) met verklaringen komen? Zijn de proponenten van presumpties altijd gevrijwaard van de bewijslast (de deontische functie van presumptie)?

C

3. Hoe kan de dialectische procedure profiteren van asymmetrische verdelingen van de bewijslast (de dialogische functie van presumptie)? 4. Welke soorten weerlegging doen presumpties teniet, en in welke evidentiële omstandigheden (voorwaarden voor weerlegging)? 5. Welke factoren maken een presumptie sterker, dan wel zwakker (concepties van sterkte)? Het proefschrift hoopt drie algemene doelen te bereiken: a) het ophelderen van begrippen die essentieel zijn in standaard besprekingen van presumptie (zoals ‘beroep op onwetendheid,’ ‘deliberatieprobleem,’ ‘evidentiële onzekerheid,’ ‘bewijslast,’ ‘asymmetrische verdeling,’ ‘sterkte,’ etc.); b) het ontwikkelen van dialoogmodellen voor presumpties (bijvoorbeeld uitzoeken welke interactionele patronen verbonden zijn met presumpties); en c) de verhouding tussen praktische en cognitieve presumpties verder exploreren.

219

CONCLU SI ON

Het proefschrift laat zien dat de twee soorten presumptie verschillende contextuele, pragmatische, dialogische en deontische functies vervullen, verschillende voorwaarden voor weerlegging hebben, en verbonden zijn met verschillende begrippen van ‘sterkte.’ Vandaar dat we praktische en cognitieve presumpties apart moeten analyseren als we onderzoek doen naar de rol van presumpties in dialogische situaties. Maar aangezien praktische en cognitieve presumpties op elkaar lijken, vergeleken met andere propositionele attitudes (zoals beweringen, assumpties of hypothesen), dient het overkoepelende begrip ‘presumptie’ toch een theoretisch doel. Hiermee in lijn bepleit ik een ‘gematigd pluralisme’ ten aanzien van presumpties.

220

Sažetak

Sažetak Prema standardnim shvaćanjima, presumpcije su opovrgljivi sudovi koji asimetrično smještaju teret dokaza, a uobičajeno je razlučiti ‘praktične’ i ‘kognitivne’ presumpcije. Praktične presumpcije su sudovi u skladu s kojima praktični djelatnici trebaju djelovati temeljem ne-epistemičkih razloga. Primjerice, ako je vjerojatnost kiše neizvjesna, Anne bi trebala izbjeći veću ne-epistemičku štetu (da pokisne), ponašati se kao da je sud „Kišit će“ istinit, te ponijeti kišobran. S druge strane, kognitive presumpcije su sudovi u skladu s kojima epistemički djelatnici trebaju djelovati temeljem epistemičkih razloga. Primjerice, ako Andy vidi da je njegova mačka na stablu, trebao bi vjerovati svojim osjetilima, zaključiti „Andyjeva mačka je na stablu,“ te iz ovoga suda izvoditi daljnje konkluzije. Unatoč cijeni usvajanja nekolicine neistinitih vjerovanja, oslanjanje na pouzdane izvore je manje štetno, tj. epistemički isplativo na duge staze. Ova disertacija doprinosi filozofiji i teoriji argumentacije oslanjajući se na nasljeđe tradicionalnih pristupa presumpcijama: rafinira tradicionalne pristupe pojašnjavajući njihove tvrdnje, rekonstruirajući dijelove koji nedostaju, te sprovodeći detaljnu konceptualnu i tipološku analizu. Pritom se usredotočuje na sljedeća pitanja: 1. Kako izgleda opravdanje presumpcije? Je li presumptivno zaključivanje svodljivo na argument iz neznanja? Služe li sve presumpcije izbjegavanju veće štete (pragmatična funkcija)? 2. U argumentiranoj raspravi, trebaju li oponenti presumpcija ponuditi argumente, ili im je također dozvoljeno ponuditi objašnjenja? Jesu li proponenti presumpcija uvijek izuzeti od tereta dokaza (deontička funkcija)? 3. Na koji način asimetrična raspodjela tereta dokaza doprinosi dijalektičkoj proceduri (dijaloška funkcija)?

C

4. Koje vrste pobijača poništavaju presumpcije, i u kojim evidencijskim okolnostima (uvjeti opovrgavanja)? 5. Koji čimbenici čine presumpciju snažnijom ili slabijom (poimanje snage presumpcije)? Disertacija ima tri opća cilja: a) rasvijetliti temeljne pojmove povezane sa standardnim shvaćanjima presumpcije (primjerice, 'argument iz neznanja', 'problem praktičnog izbora', 'evidencijska naizvjesnost', 'teret dokaza', 'asimetrična raspodjela', 'snaga', itd.); b) izraditi dijaloški model presumpcije (primjerice, saznati koji se obrasci interakcije uistinu povezuju s presumpcijama); te c) unaprijediti istraživanje odnosa između praktičnih i kognitivnih presumpcija. Nastojim pokazati da dvije vrste presumpcija imaju drukčije kontekstualne, pragmatične, dijaloške, i deontičke funkcije; različite uvjete opovrgavanja, te da povlače

221

CONCLU SI ON

drukčija poimanja 'snage'. Stoga, dok god istražujemo njihove primjene u dijaloškim kontekstima, praktične i kognitivne presumpcije zahtjevaju odvojenu analizu. No budući da praktične i kognitive presumpcije u mnogočemu nalikuju kada ih usporedimo s drugim propozicijskim stavovima (primjerice, tvrdnjama, pretpostavkama, ili hipotezama), pojam 'presumpcije' i dalje je u ograničenoj mjeri teorijski svrhovit. U skladu s time, argumentiram u prilog 'umjerenog pluralizma' presumpcija.

222

Curriculum Vitae

Curriculum Vitae Petar Bodlović was born on 31st October 1988 in Dubrovnik, Croatia. He studied philosophy and sociology at the University of Zagreb. In his B.A. thesis, he argued that the Principle of Charity helps resolve the ambiguity of conclusion by prescribing a benevolent interpretation of the argument’s structure. During graduate study, he worked as a Junior Assistant at the course “Strategies of Informal Reasoning—Critical Thinking.” In his M.A. thesis, defended in 2013, he critically evaluated Douglas Walton’s claim that the argument’s validity is context-dependent. After starting his Ph.D. at the University of Granada (Spain), he was awarded a scholarship at the University of Groningen. From 2017, he has been investigating the role of presumptions in argumentative dialogues under the supervision of dr. Jan Albert van Laar and prof. dr. Barteld Kooi. So far, he has published seven scientific articles in peer-reviewed journals and several papers in student journals and conference proceedings. In addition, he presented his work at twenty conferences or workshops and won “Frans van Eemeren Prize” (2019) and “J. Anthony Blair Prize” (2020) for outstanding student papers. Besides research, he likes music, football (soccer), literature, long walks, and good company.

C

223

CONCLU SI ON

224

References

References Aijaz, I., McKeown-Green, & A. Webster. 2013. Burdens of proof and the case for unevenness. Argumentation 27(3): 259-282. Allen, M. 2004. Smart thinking (2nd edition). Melbourne: Oxford University Press. Allen, R.J. 1980. Presumptions in civil actions reconsidered. Iowa Law Review 66: 843-867. Allen, R.J. 2014. Burdens of proof. Law, Probability and Risk 3(3-4): 195-219. Allhoff, F. 2014. Risk, Precaution, and Nanotechnology. In In Pursuit of Nanoethics, eds. B. Gordijn, & A.M. Cutter, 107-131. Dordrecht: Springer Science+Business Media. Archbold, J.F. 1831. A Summary of the Law Relative to Pleading in Criminal Cases. London: Printed for R. Pheney. Atkinson, K., T. Bench-Capon, & D. Walton. 2013. Distinctive features of persuasion and deliberation dialogues. Argument and Computation 4(2): 105-127. Atkinson, K., T. Bench-Capon, F.J. Bex, T.F. Gordon, H. Prakken, G. Sartor, & B. Verheij. 2020. In Memoriam Douglas N. Walton: The influence of Doug Walton on AI and Law. Artificial Intelligence and Law 28(3): 281-326. Austin, J.L. 1962. How to do Things with Words (2nd edition), eds. J. O. Urmson, & M. Sbisá. Cambridge, MA: Harvard University Press. Bermejo-Luque, L. 2011. Giving reasons. A Linguistic-Pragmatic Approach to Argumentation Theory. Dordrecht: Springer Science+Business Media. Bermejo-Luque, L. 2016. Being a correct presumption vs. being presumably the case. Informal Logic 36(1): 1-25. Bermejo-Luque, L. 2019. On the Relationship between Presumptions and Burdens of Proof. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo-Luque, 244-257. Tuscaloosa: The University of Alabama Press.

C

Blair, J.A. 2012. Argumentation as Rational Persuasion. Argumentation 26: 71-81. Bodlović, P. 2017. Dialogical Features of Presumptions: Difficulties for Walton’s New Dialogical Theory. Argumentation 31(3): 513-534. Bodlović, P. 2019a. What determines the strength of practical and cognitive presumptions? In Proceedings of the Ninth Conference of the International Society for the Study of Argumentation (ISSA), eds. B. Garssen, D. Godden, G.R. Mitchell, & J.H.M. Wagemans, 106-117. Amsterdam: Sic Sat. Bodlović, P. 2019b. Presumptions, and How They Relate to Arguments from Ignorance. Argumentation 33: 579-604. Bodlović, P. 2020a Structural differences between practical and cognitive presumptions. In Proceedings of the third European Conference on Argumentation (ECA) Reason to dissent (VOL 1), eds. C. Dutilh Novaes, H. Jansen, J.A. van Laar, & B. Verheij, 73-94. Bodlović, P. 2020b. On Presumptions, Burdens of Proof, and Explanations. Informal Logic 40(2): 255-294. Bodlović, P. 2020c. Presumptions, burdens of proof, and explanations. OSSA Conference Archive 17: 1-22.

225

CONCLU SI ON

Bodlović, P. 2021. On the differences between practical and cognitive presumptions. Argumentation 35: 287-320. Bodlović, P. On Presumptions, and Asymmetrical Allocations of Burden of Proof. Forthcoming. Brown, R. 1970. The Burden of Proof. American Philosophical Quarterly 7(1): 74-82. Capper, C., & R. Cross. 1990. Cross on Evidence. London: Butterworths. Cargile, J. 1997. On the Burden of Proof. Philosophy 72(279): 59-83. Carnap, R. 1947. Meaning and necessity. Chicago: University of Chicago Press. Chalmers, D.J. 2020. What is conceptual engineering and what should it be? Inquiry, forthcoming, DOI: 10.1080/0020174X.2020.1817141. Cleary, E.W. 1959. Presuming and pleading: An essay on juristic immaturity. Stanford Law Review 12(1): 5-28. CNBC.com staff. 2016. Trump tweets new voter fraud claims, blasts CNN, but offers no evidence. URL accessed 9 November 2019: https://www.cnbc.com/2016/11/28/trump-tweets-new-voter-fraudclaims-blasts-cnn-but-offers-no-evidence.html. Cohen, L.J. 1992. An essay on belief and acceptance. Oxford: Clarendon Press. Copi, I.M., C. Cohen, & K. McMahon. 2016. Introduction to Logic (14th edition). London and New York: Routledge. Corredor, C. 2017. Presumptions in Speech Acts. Argumentation 31: 573-589. Cummings, L. 2002. Reasoning Under Uncertainty: The Role of Two Informal Fallacies in an Emerging Scientific Inquiry. Informal Logic 22(2): 113-136. Daly, C. 2010. An Introduction to Philosophical Methods. Toronto: Broadview guides to philosophy. Dare, T., & J. Kingsbury. 2008. Putting the Burden of Proof in Its Place: When Are Differential Allocations Legitimate? The Southern Journal of Philosophy XVI: 503-519. Dutilh Novaes, C., & E. Reck. 2017. Carnapian explication, formalisms as cognitive tools, and the paradox of adequate formalization. Synthese 194: 195-215. Elgin, C.Z. 2010. Persistent disagreement. In Disagreement, eds. R. Feldman, & T. Warfield, 53-68. Oxford: Oxford University Press. Feldman, R. 2003. Epistemology. Upper Saddle River, New Jersey: Prentice Hall. Flew, A. 1972. The Presumption of Atheism. Canadian Journal of Philosophy 2(1): 29-46. Freeman, J.B. 2005. Acceptable premises. Cambridge: Cambridge University Press. Freeman, J.B. 2019. Factors for Evaluating Presumptions and Presumptive Inferences. Argumentation, https://doi.org/10.1007/s10503-018-9468-8. Freeman, J.B. 2020. Commentary on Bodlović’s Structural differences between practical and cognitive presumptions. In Proceedings of the third European Conference on Argumentation (ECA) Reason to dissent (VOL 1), eds. C. Dutilh Novaes, H. Jansen, J.A. van Laar, & B. Verheij, 95-101. Gama, R. 2017. The Nature and the Place of Presumptions in Law and Legal Argumentation. Argumentation 31(3): 555-572.

226

References

Giddings, F.H. 1922. The Grounds of Presumption. The Journal of Philosophy 19(23): 617-624. Glock, H.J. 2008. What Is Analytic Philosophy? New York: Cambridge University Press. Godden, D. 2011. Presumptions in argument: Epistemic versus social approaches. In Argumentation: cognition and Community. Proceedings of the 9th International Conference of the Ontario Society for the Study of Argumentation (OSSA), ed. F. Zenker, 1-13. Windsor, ON (CD ROM). Godden, D. 2017. Presumption as a Modal Qualifier: Presumption, Inference, and Managing Epistemic Risk. Argumentation 31(3): 485-511. Godden, D. 2019. Analyzing Presumption as a Modal Qualifier. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo-Luque, 206-220. Tuscaloosa: The University of Alabama Press. Godden, D. 2020. Commentary on Petar Bodlović: Presumptions, burdens of proof, and explanations. OSSA Conference Archive 18: 1-7. Godden, D., & D. Walton. 2007. A theory of presumption for everyday argumentation. Pragmatics & Cognition 15(2): 313-346. Goldman, A. 1976. Discrimination and Perceptual Knowledge. Journal of Philosophy 73(20): 771-791. Goldman, A. 1994. Argumentation and Social Epistemology. Journal of Philosophy 91(1): 27-49. Goldman, A. 2003. An Epistemological Approach to Argumentation. Informal Logic 23(1): 51-63. Goodnight, G.T. 2019. The Liberal-Progressive and Conservative Presumptions: On Deliberation, Debate, and Public Argument. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo-Luque, 156-180. Tuscaloosa: The University of Alabama Press. Goodwin, J. 2000. The noncooperative pragmatics of arguing. In Pragmatics in 2000: Selected papers from the 7th International Pragmatics Conference (Vol. 2), ed. Eniko Nemeth, 263-277. Antwerp: International Pragmatics Association. Goodwin, J. 2001. One Question, Two Answers. OSSA Conference Archive: 1-16.

C

Goodwin, J. 2007. Argument Has No Function. Informal Logic 27(1): 69-90. Govier, T. 2010. A Practical Study of Argument (7th edition). Wadsworth, Cengage Learning. Grice, H.P. 1975. Logic and Conversation. In Syntax and Semantics: Speech Acts (Vol. 3), eds. P. Cole, & J.L. Morgan, 41-58. New York: AcademicPress. Hahn, U., & M. Oaksford. 2007. The burden of proof and its role in argumentation. Argumentation 21(1): 39-61. Hamblin, C. 1970. Fallacies. London: Methuen & Co Ltd. Hansen, H.V. 2003. Theories of presumption and burden of proof. In Proceedings of the Ontario Society for the Study of Argumentation (OSSA), eds. J.A. Blair et al., 1-12. Windsor, ON: Ontario Society for the Study of Argumentation (OSSA). Hansen, H.V. 2019. Historical selections. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo-Luque, 5-19. Tuscaloosa: The University of Alabama Press.

227

CONCLU SI ON

Hasan, A., & R. Fumerton. 2016. Foundationalist Theories of Epistemic Justification. In  The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), ed. E.N. Zalta, URL=https://plato.stanford.edu/ archives/win2016/entries/justep-foundational/, 22 May 2018. Houtlosser, P. 2003. Commentary on H.V. Hansen’s ‘Theories of presumption and burden of proof’. In Proceedings of the Ontario Society for the Study of Argumentation (OSSA), eds. J.A. Blair et al., 1-4. Windsor, ON: Ontario Society for the Study of Argumentation (OSSA). Ilbert, C.P. 2019. On Presumption and Burden of Proof. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo-Luque, 83-87. Tuscaloosa: The University of Alabama Press. Johnson, R.H. 2000. Manifest rationality: a pragmatic theory of argument. Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc. Kauffeld, F.J. 1998. Presumptions and the Distribution of Argumentative Burdens in Acts of Proposing and Accusing. Argumentation 12(2): 245-266. Kauffeld, F.J. 2003. The Ordinary Practice of Presuming and Presumption with Special Attention to Veracity and the Burden of Proof. In Anyone Who has a View: Theoretical Contributions to the Study of Argumentation, eds. F.H. van Eemeren et al., 133-146. Dordrecht: Springer. Kauffeld, F.J. 2005. Presumptions and Shifting the Burden of Proof. Paper Presented at the IPrA Conference. http://www2.arnes .si/~ffljz agar/Kauff eld_paper .pdf. 10 October, 2019. Kauffeld, F. 2009. Presuming and Presumption in Everyday Argumentation: A response to Godden and Walton. In Argument Cultures: Proceedings of OSSA 09, ed. J. Ritola, 1-13. Windsor, ON: OSSA. Kauffeld, F. 2019. A Rhetorically Oriented Account of Presumption and Probative Obligations in Normative Pragmatic Terms. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo-Luque, 257-273. Tuscaloosa: The University of Alabama Press. Kelly, T. 2005. Moorean facts and belief revision, or can the skeptic win? Philosophical Perspectives 19: 179-209. Kokott, J. 1998. The Burden of Proof in Comparative and International Human Rights Law: Civil and Common Law Approaches with Special Reference to the American and German Legal Systems. The Hague: Kluwer Law International. Koplin, J., & M. Selgelid. 2015 . Burden of proof in bioethics. Bioethics 29(9): 597-603. Kornblith, H. 2016. Philosophical Naturalism. In The Oxford Handbook of Philosophical Methodology, eds. H. Cappelen, T.S. Gendler, & J. Hawthorne, 147-159. Oxford: Oxford University Press. Krabbe, E.C.W. 1995. Appeal to Ignorance. In Fallacies: classical and contemporary readings, eds. H.V. Hansen, & R.C. Pinto, 251-265. University Park, Pennsylvania: The Pennsylvania State University Press. Krabbe, E.C.W. 2001. Strategies in dialectic and rhetoric. In Argumentation and its applications: Proceedings from the Conference of the Ontario Society for the Study of Argumentation, eds. H.V. Hansen, C.W. Tindale, J.A. Blair, R.H. Johnson, & R.C. Pinto, 1-13. Windsor, ON: University of Windsor. Krabbe, E.C.W., & J.A. van Laar. 2011. The Ways of Criticism. Argumentation 25: 199-227. Krabbe, E.C.W., & J.A. van Laar. 2021. In the quagmire of quibbles: a dialectical exploration. Synthese 198: 3459-3476. 228

References

Leite, A. 2005. A localist solution to the regress of epistemic justification. Australasian Journal of Philosophy 83(3): 395-421. Lewiński, M.. 2017. Argumentation Theory Without Presumptions. Argumentation 31(3): 591-613. Locke, J. 1995. Four Sorts of Arguments. In Fallacies: classical and contemporary readings, eds. H.V. Hansen, & R.C. Pinto, 55-57. University Park, Pennsylvania: The Pennsylvania State University Press. Lycan, W.G. 2003. Free will and the burden of proof. Royal Institute of Philosophy Supplement 53: 107-122. Macagno, F. 2010. Dialectical and heuristic arguments: presumptions and burden of proof. In Dialectics, dialogue and argumentation. An examination of Douglas Walton’s theories of reasoning, eds. C. Reed, & C. Tindale, 45-57. London: College Publications. Macagno, F. 2012. Presumptive Reasoning in Interpretation. Implicatures and Conflicts of Presumptions. Argumentation 26: 233-265. Macagno, F., & G. Damele. 2013. The Dialogical Force of Implicit Premises: Presumptions in Enthymemes. Informal Logic 33(3): 361-389. Macagno, F., & D. Walton. 2011. Reasoning from paradigms and negative evidence. Pragmatics & Cognition 19(1): 92-116. Macagno, F., & D. Walton. 2012. Presumptions in legal argumentation. Ratio Juris 25(3): 271-300. Moore, G.E. 2000. Four Forms of Skepticism. In Epistemology: An Anthology, eds. E. Sosa, & J. Kim, 2729. Oxford: Blackwell Publishers. Neville-Shepard, R. 2019. Post-presumption argumentation and the post-truth world: on the conspiracy rhetoric of Donald Trump. Argumentation and Advocacy 55(3): 175-193. Pascal, B. 2004. Pensées (edited and translated by Roger Ariew). Indianapolis/Cambridge: Hackett Publishing Company. Perelman, C., & L. Olbrechts-Tyteca. 1969. The New Rhetoric. A Treatise on Argumentation. Notre Dame: University of Notre Dame Press. Petroski, K. 2008. The public face of presumptions. Episteme 5(3): 388-401.

C

Pigliucci, M., & M. Boudry. 2014. Prove it! The Burden of Proof Game in Science vs. Pseudoscience Disputes. Philosophia 42(2): 487-502. Pinto, R.C. 2001. Argument, inference and dialectic: collected papers on informal logic (with an introduction by Hans V. Hansen). Dordrecht: Kluwer. Plato. 2000. The Republic (edited by G.R.F. Ferrari and translated by T. Griffith). Cambridge: Cambridge University Press. Plumer, G. 2017. Presumptions, Assumptions, and Presuppositions of Ordinary Arguments. Argumentation 31(3): 469-484. Pollock, J.L. 1987. Defeasible reasoning. Cognitive science 11: 481-518. Prakken, H. 2004. Analysing reasoning about evidence with formal models of argumentation. Law, Probability & Risk 3: 33-50. Prakken, H., C. Reed, & D. Walton. 2005. Dialogues about the burden of proof. In Proceedings of the Tenth International Conference on Artificial Intelligence and Law, 115-124. New York: The Association for Computing Machinery Press. 229

CONCLU SI ON

Prakken, H., & G. Sartor. 2006. Presumptions and Burdens of Proof. In Legal Knowledge and Information Systems: JURIX 2006: The Nineteenth Annual Conference, ed. T.M. van Engers, 21-30. Amsterdam: IOS Press. Prakken, H., & G. Sartor. 2009. A Logical Analysis of Burdens of Proof. In Legal Evidence and Proof: Statistics, Stories, Logic, eds. H. Kaptein, H. Prakken, & B. Verheij, 223-253. Farnham: Ashgate Publishing (Applied Legal Philosophy Series). Presumption. In A Dictionary of Modern English Usage, Cambridge University Press, https://dictionary. cambridge.org/dictionary/english/presumption. Accessed 16 March 2021. Presumption. In Cambridge Advanced Learner’s Dictionary, Cambridge University Press, https:// dictionary.cambridge.org/dictionary/english/presumption. Accessed 16 March 2021. Presumption. In Cambridge Online English Dictionary, Cambridge University Press, https://dictionary. cambridge.org/dictionary/english/presumption. Accessed 16 March 2021. Presumption. In Merriam-Webster.com Dictionary, Merriam-Webster, https://www.merriam-webster.com/ dictionary/presumption. Accessed 1 May 2021. Räikkä, J. 1997. Burden of Proof Rules in Social Criticism. Argumentation 11: 463-477. Räikkä, J. 2005. Global Justice and the Logic of the Burden of Proof. Metaphilosophy 36(1/2): 228-239. Raz, J. 2009. Reasons: practical and adaptive. In Reasons for Action, eds. D. Sobel, & S. Wall, 37-57. Cambridge: Cambridge University Press. Reid, T. 1785. Essays on the Intellectual Powers of Man. Edinburgh: John Bell. Rescher, N. 1976. Plausible Reasoning. Assen/Amsterdam: Van Gorcum. Rescher, N. 1977. Dialectics: A controversy-oriented approach to the theory of knowledge. Albany, NY: SUNY Press. Rescher, N. 2006. Presumption and the practices of tentative cognition. Cambridge: Cambridge University Press. Rescher, N. 2007. Dialectics. A Classical Approach to Inquiry. Frankfurt: Ontos Verlag. Rescorla, M. 2009a. Shifting the burden of proof? Philosophical Quarterly 59(234): 86-109. Rescorla, M. 2009b. Epistemic and dialectical regress. Australasian Journal of Philosophy 87(1), 43-60. Reynolds, W. 1897. The Theory of the Law of Evidence as Established in the United States. Chicago: Callaghan and Company. Rhode, C. 2017. The Burden of Proof in Philosophical Persuasion Dialogue. Argumentation 31(3): 535-554. Scott, D.N. 2005. Shifting the Burden of Proof: The Precautionary Principle and Its Potential for the “Democratization” of Risk. In Law and Risk, ed. Law Commission of Canada, 50-86. Vancouver/ Toronto: UBCPress. Scriven, M. 1966. Primary philosophy. New York: McGraw-Hill. Searle, J., & D. Vanderveken. 1985. Foundations of Illocutionary Logic. Cambridge: Cambridge University Press. Senor, T.D. 1996. The Prima/Ultima Facie Justification Distinction in Epistemology. Philosophy and Phenomenological Research 56(3): 551-566.

230

References

Soukhanov, A.H. 1992. The American Heritage Dictionary of the English Language (3rd edition). Houghton Mifflin Harcourt: University of Michigan. Strong, J.W. 1992. McCormick on Evidence (4th edition). St Paul, Minnesota: West Publishing Co. Thayer, J.B. 2019. The Burden of Proof. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo Luque, 7483. Tuscaloosa: The University of Alabama Press. Tindale, C.W. 2007. Fallacies and Argument Appraisal. New York: Cambridge University Press. Tseronis, A. 2009. Qualifying Standpoints. Stance adverbs as a presentational device for managing the burden of proof (doctoral thesis). Utrecht: LOT. Turri, J., M. Alfano, & J. Greco. 2019. Virtue Epistemology. In The Stanford Encyclopedia of Philosophy (Fall 2019 Edition), ed. Edward N. Zalta. https://plato.stanford.edu/archives/fall2019/entries/ epistemology-virtue/. 12 February 2020. Tuzet, G. 2015. On the absence of evidence. In Argument types and fallacies in legal argumentation, eds. T. Bustamante, & C. Dahlman, 37-53. Cham: Springer. Ullmann-Margalit, E. 1983. On presumption. Journal of Philosophy 80(3): 143-163. Ullmann-Margalit, E. 2000. On Not Wanting to Know. In Reasoning practically, ed. E. Ullmann-Margalit, 72-85. New York: Oxford University Press. Ullmann-Margalit, E., & A. Margalit. 1982. Analyticity by Way of Presumption. Canadian Journal of Philosophy 12(3): 435-452. Ullmann-Margalit, E., & A. Margalit. 1992. Holding true and holding as true. Synthese 92: 167-187. van Eemeren, F.H., B. Garssen, E.C.W. Krabbe, A.F. Snoeck Henkemans, B. Verheij, & J.H.M. Wagemans. 2014. Handbook of Argumentation Theory. Dordrecht: Springer Science+Business Media. van Eemeren, F.H., B. Garssen, & B. Meuffels. 2009. Fallacies and Judgments of Reasonableness. Springer Science+Business Media B.V. van Eemeren, F.H., & R. Grootendorst. 2004. A Systematic Theory of Argumentation. The PragmaDialectical Approach. Cambridge: Cambridge University Press.

C

van Eemeren, F.H., R. Grootendorst, & A.F. Snoeck Henkemans. 2002. Argumentation: Analysis, Evaluation, Presentation. Mahwah, New Jersey/London: Lawrence Erlbaum Associates. van Eemeren, F.H., & P. Houtlosser. 2002. Strategic Maneuvering with the burden of proof. In Advances in Pragma-Dialectics, ed. F.H. van Eemeren, 13-29. Sic Sat/Amsterdam. van Eemeren, F.H., & P. Houtlosser. 2015. The Case of Pragma-Dialectics (Chapter 8). In Reasonableness and Effectiveness in Argumentative Discourse, ed. F.H. van Eemeren, 149-181). Switzerland: Springer International Publishing. van Laar, J.A. 2014. Criticism in Need of Clarification. Argumentation 28: 401-423. van Laar, J.A., & Krabbe, E.C.W. 2013. The Burden of Criticism: Consequences of Taking a Critical Stance. Argumentation 27(2): 201-224 Walton, D. 1988. Burden of Proof. Argumentation 2(2): 233-254.

231

CONCLU SI ON

Walton, D. 1990. What is reasoning? What is an argument? The Journal of Philosophy 87(8): 399-419. Walton, D. 1992. Nonfallacious arguments from ignorance. American Philosophical Quarterly 29(4): 381-387. Walton, D. 1993. The speech act of presumption. Pragmatics & Cognition 1(1): 125-148. Walton, D. 1996a. Argumentation schemes for presumptive reasoning. Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc. Walton, D. 1996b. Argument Structure: A Pragmatic Theory. Toronto: University of Toronto Press. Walton, D. 1998. The new dialectic: Conversational contexts of argument. Toronto: University of Toronto Press. Walton, D. 1999. Informal fallacy. In The Cambridge Dictionary of Philosophy (2nd edition), ed. R. Audi, 431-435. New York: Cambridge University Press. Walton, D. 2001. Abductive, presumptive and plausible arguments. Informal Logic 21(2): 141-169. Walton, D. 2006. Fundamentals of Critical Argumentation. New York: Cambridge University Press. Walton, D. 2008a. Presumption, burden of proof and lack of evidence. L’Analisi Linguistica e Latteraria 16(1): 49-71. Walton, D. 2008b. A dialogical theory of presumption. Artificial Intelligence and Law 16(2): 209-243. Walton, D. 2014. Burden of proof, presumption and argumentation. New York: Cambridge University Press. Walton, D. 2019. The Speech Act of Presumption by Reversal of Burden of Proof. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo-Luque, 220-234. Tuscaloosa: The University of Alabama Press. Walton, D., & E.C.W. Krabbe. 1995. Commitment in dialogue. Basic concepts of interpersonal reasoning. Albany: Suny Press. Walton, D., C. Reed, & F. Macagno. 2008. Argumentation schemes. New York: Cambridge University Press. Whately, R. 1963. Elements of Rhetoric (7th ed.). Carbondale IL: Southern Illinois University Press. Whately, R. 2019. Presumptions and Burden of Proof. In Presumptions and burdens of proof. An anthology of argumentation and the law, eds. H.V. Hansen, F.J. Kauffeld, J.B. Freeman, & L. Bermejo-Luque, 45-61. Tuscaloosa: The University of Alabama Press. Wigmore, J. H. 1962. Evidence in Trials at Common Law (Volume IX). New York: Little, Brown and Company. Williamson T. 2011. Philosophical expertise and the burden of proof. Metaphilosophy 42(3): 215-229. Witek, M. 2019. Illocution and accommodation in the functioning of presumptions. Synthese, https://doi. org/10.1007/s11229-019-02459-4. Woods, J., A. Irvine, & D. Walton. 2004. Argument: critical thinking, logic and the fallacies, 2nd edition. Toronto: Pearson/Prentice Hall. Woods, J., & D. Walton.1978. The Fallacy of 'Ad Ignorantiam'. Dialectica 32(2): 87-99. Wreen, M. 1989. Light from Darkness, From Ignorance Knowledge. Dialectica 43(4): 299-314.

232

References

C

233