Explanation, Misc

In a recent paper, Erasmus et al. (2021) defend the idea that the ambiguity of the term “explanation” in explainable AI (XAI) can be solved by adopting any of four different extant accounts of explanation in the philosophy of science: the Deductive Nomological, Inductive Statistical, Causal Mechanical, and New Mechanist models. In this chapter, I show that the authors’ claim that these accounts can be applied to deep neural networks as they would to any natural phenomenon is mistaken. I also (...) provide a more general argument as to why the notion of explainability as it is currently used in the XAI literature bears little resemblance to the traditional concept of scientific explanation. It would be more fruitful to use the label “understandable AI” to avoid the confusion that surrounds the goal and purposes of XAI. In the second half of the chapter, I argue for a pragmatic conception of understanding that is better suited to play the central role attributed to explanation in XAI. Following De Regt (2017) and Kuorikoski and Ylikoski (2015), the conditions of satisfaction for understanding an ML system are fleshed out in terms of an agent’s success in using the system. (shrink)

The philosophical literature on scientific explanation in neuroscience has been dominated by the idea of mechanisms. The mechanist philosophers often claim that neuroscience is in the business of finding mechanisms. This view has been challenged in numerous ways by showing that there are other successful and widespread explanatory strategies in neuroscience. However, the empirical evidence for all these claims was hitherto lacking. Empirical evidence about the pervasiveness and uses of various explanatory strategies in neuroscience is particularly needed because examples and (...) case studies that are used to illustrate philosophical claims so far tend to be hand-picked. The risk of confirmation bias is therefore considerable: when looking for white swans, all one finds is that swans are white. The more systematic quantitative and qualitative bibliometric study of a large body of relevant literature that we present in this paper can put such claims into perspective. Using bibliometric tools, we identify the typical linguistic patterns used in the alleged mechanistic, dynamical, and topological explanations in the literature, their preponderance and how they change over time. Our findings show abundant use of mechanistic language, but also the presence of a significant neuroscience literature using topological and dynamical explanatory language, which grows over time and increasingly differentiates from each other and from mechanistic explanations. (shrink)

The bridges of Königsberg case has been widely cited in recent philosophical discussions on scientific explanation as a potential example of a structural explanation of a physical phenomenon. However, when discussing this case, different authors have focused on two different versions, depending on what they take the explanandum to be. In one version, the explanandum is the _failure_ of a given individual in performing an Eulerian walk over the bridge system. In the other version, the explanandum is the _impossibility_ of (...) performing an Eulerian walk over the bridges. The goal of this paper is to show that only the latter version amounts to a real case of a structural explanation. I will also suggest how to fix the first version, and show how my remarks apply to other purported cases of structural explanations. (shrink)

In response to widespread use of automated decision-making technology, some have considered a right to explanation. In this paper I draw on insights from philosophical work on explanation to present a series of challenges to this idea, showing that the normative motivations for access to such explanations ask for something difficult, if not impossible, to extract from automated systems. I consider an alternative, outcomes-focused approach to the normative evaluation of automated decision-making, and recommend it as a way to pursue the (...) goods originally associated with explainability. (shrink)

We develop and apply a multi-dimensional conception of explanatory depth towards a comparative analysis of inflationary and bouncing paradigms in primordial cosmology. Our analysis builds on earlier work due to Azhar and Loeb (2021) that establishes initial condition fine-tuning as a dimension of explanatory depth relevant to debates in contemporary cosmology. We propose dynamical fine-tuning and autonomy as two further dimensions of depth in the context of problems with instability and trans-Planckian modes that afflict bouncing and inflationary approaches respectively. In (...) the context of the latter issue, we argue that the recently formulated trans-Planckian censorship conjecture leads to a trade-off for inflationary models between dynamical fine-tuning and autonomy. We conclude with the suggestion that explanatory preference with regard to the different dimensions of depth is best understood in terms of differing attitudes towards heuristics for future model building. (shrink)

Many AI systems that make important decisions are black boxes: how they function is opaque even to their developers. This is due to their high complexity and to the fact that they are trained rather than programmed. Efforts to alleviate the opacity of black box systems are typically discussed in terms of transparency, interpretability, and explainability. However, there is little agreement about what these key concepts mean, which makes it difficult to adjudicate the success or promise of opacity alleviation methods. (...) I argue for a unified account of these key concepts that treats the concept of understanding as fundamental. This allows resources from the philosophy of science and the epistemology of understanding to help guide opacity alleviation efforts. A first significant benefit of this understanding account is that it defuses one of the primary, in-principle objections to post hoc explainable AI (XAI) methods. This “rationalization objection” argues that XAI methods provide mere rationalizations rather than genuine explanations. This is because XAI methods involve using a separate “explanation” system to approximate the original black box system. These explanation systems function in a completely different way than the original system, yet XAI methods make inferences about the original system based on the behavior of the explanation system. I argue that, if we conceive of XAI methods as idealized scientific models, this rationalization worry is dissolved. Idealized scientific models misrepresent their target phenomena, yet are capable of providing significant and genuine understanding of their targets. (shrink)

Extra-mathematical explanations explain natural phenomena primarily by appeal to mathematical facts. Philosophers disagree about whether there are extra-mathematical explanations, the correct account of them if they exist, and their implications (e.g., for the philosophy of scientific explanation and for the metaphysics of mathematics) (Baker 2005, 2009; Bangu 2008; Colyvan 1998; Craver and Povich 2017; Lange 2013, 2016, 2018; Mancosu 2008; Povich 2019, 2020; Steiner 1978). In this discussion note, I present three desiderata for any account of extra-mathematical explanation and argue (...) that Baron’s (2020) U-Counterfactual Theory fails to meet each of them. I conclude with some reasons for pessimism that a successful account will be forthcoming. (shrink)

In her new book Reconstructing Reality (henceforth RR), Margaret Morrison’s main target is the kind of information about the world (or, more specifically, about physical and biological systems) one can extract from the ‘reconstructive methods and practices of science’ (p. 1). To address this, Morrison focuses on three central kinds of interrelated strategies for ‘recasting nature’ (p. 2) by using reconstructive methods and practices: (i) abstract mathematical explanations and understanding (Part 1 of the book), (ii) scientific models (Part 2), and (...) (iii) computer simulations (Part 3). Hence, RR is the ambitious attempt to connect and analyse three major topics in current philosophy of science. (shrink)

This paper introduces a new kind of explanation that we describe as ‘purely theoretical’. We first present an example, E, of what we take to be a case of purely theoretical explanation. We then show that the explanation we have in mind does not fit neatly into any of the existing categories of explanation. We take this to give us prima facie motivation for thinking that purely theoretical explanation is a distinctive kind of explanation. We then argue that it can (...) earn its keep via application to two existing literatures: the literature on how we explain the truth of true negative existential propositions and the literature on how we explain the truth of true propositions about the past. We reply to some possible concerns regarding the introduction of purely theoretical explanations. We conclude that there is nothing obviously wrong with them and explore the ramifications for particular debates in metaphysics. (shrink)

Several have denied that there is, specifically, a criterion of identity for persons and some deny that there are, for any kind, diachronic criteria of identity. I argue, however, that there are no criteria of identity, either synchronic or diachronic, for any kind whatsoever. I begin by elaborating the notion of a criterion of identity in order to clarify what exactly is being denied when I maintain there are none. I examine the motivation of those who qualify in some way (...) the general claim that there are synchronic and diachronic criteria of identity for every kind, then present my direct and categorical argument against such criteria. I next evaluate the objections of those who argue that rejecting criteria of identity has untenable results. These objections are ineffective, each based on the incorrect assumption that if there is no criterion of identity for a kind, the identity of an instance of that kind is independent of its qualities. I conclude by considering some of the upshots of rejecting criteria of identity and the insight doing so provides into things in general and the limits of ontological inquiry. (shrink)

A brief outline for some alternative lines of thought on the general structure of how-possible reasoning and its implications for formally and informally conceivable things, and the concept of mathematical truth.

I defend an interpretation of Aristotle’s Posterior Analytics Book I which distinguishes between two projects in different passages of that work: (i) to explain what a given science is and (ii) to explain what properly scientific knowledge is. I present Aristotle’s theory in answer to ii, with special attention to his definition of scientific knowledge in 71b9-12 and showing how this is developed on chapters I.2-9 and I.13 into a solid Theory of Scientific Demonstration. The main point of this theory (...) is that demonstrations need to capture relevant explanations. Some formal requirements of the demonstration (as the syllogistic structure and coextension between terms) are unfoldings of the main project, i.e., to capture and present properly relevant causal-explanatory relations. (shrink)

Batterman and Rice ([2014]) argue that minimal models possess explanatory power that cannot be captured by what they call ‘common features’ approaches to explanation. Minimal models are explanatory, according to Batterman and Rice, not in virtue of accurately representing relevant features, but in virtue of answering three questions that provide a ‘story about why large classes of features are irrelevant to the explanandum phenomenon’ ([2014], p. 356). In this article, I argue, first, that a method (the renormalization group) they propose (...) to answer the three questions cannot answer them, at least not by itself. Second, I argue that answers to the three questions are unnecessary to account for the explanatoriness of their minimal models. Finally, I argue that a common features account, what I call the ‘generalized ontic conception of explanation’, can capture the explanatoriness of minimal models. (shrink)

_ Source: _Volume 10, Issue 3, pp 441 - 456 Several kinds of historical alternatives are distinguished. Different kinds of historical alternatives are valuable to the practice of history for different reasons. Important uses for historical alternatives include representing different sides of historical disputes; distributing chances of different outcomes over alternatives; and offering explanations of why various alternatives did _not_ in fact happen. Consideration of counterfactuals about what would have happened had things been different in particular ways plays particularly useful (...) roles in reasoning about historical analogues of current conditions; reasoning about causal claims; and in evaluating historical explanations. When evaluating the role of alternative histories in historical thinking, we should keep in mind the uses of historical alternatives that go well beyond the long-term and specific scenarios that are the focus of so-called “counterfactual history”. (shrink)

Theories of explanation seek to tell us what distinctively explanatory information is. The most ambitious ones, such as the DN-account, seek to tell us what an explanation is, tout court. Less ambitious ones, such as causal theories, restrict themselves to a particular domain of inquiry. The least ambitious theories constitute outright skepticism, holding that there is no reasonably unified phenomenon to give an account of. On these views, it is impossible to give any theories of explanation at all. I argue (...) that both the less ambitious and outright skeptical varieties are committed to a certain context-sensitivity of our explanatory discourse. And though this discourse is almost certainly context-sensitive in some respects, it does not exhibit the context-sensitivity less than fully ambitious theories are committed to. Therefore, all accounts that seek to restrict themselves in scope, including causal accounts of explanation, fail. (shrink)

Stathis Psillos divides his account into three sections: causation, laws of nature, and explanation. He begins the causation section with Hume's classic "reductive" account and then focuses on the subsequent division between Humean and non-Humean accounts, examining topics such as regularities and singular causation, causation and counterfactuals, and causation and mechanism.

This book is based on previously unpublished lectures that Wisdom delivered at the University of Virginia. Its content goes significantly beyond that of his other books. Here he is concerned with how misunderstandings about what it is to prove something or what it is to explain something can infect our thinking in many different fields.

argumentation has been shown to be a powerful tool within many fields such as artificial intelligence, logic and legal reasoning. In this paper we enhance Dung’s well-known abstract argumentation framework with explanatory capabilities. We show that an explanatory argumentation framework (EAF) obtained in this way is a useful tool for the modeling of scientific debates. On the one hand, EAFs allow for the representation of explanatory and justificatory arguments constituting rivaling scientific views. On the other hand, different procedures for selecting (...) arguments, corresponding to different methodological and epistemic requirements of theory evaluation, can be formulated in view of our framework. (shrink)

Metaphysical theories heavily rely on the use of primitives to which they typically appeal. I will start by examining and evaluating some traditional well-known theories and I will discuss the role of primitives in metaphysical theories in general. I will then turn to a discussion of claims of between theories that, I think, depend on equivalences of primitives, and I will explore the nature of primitives. I will then claim that almost all explanatory power of metaphysical theories comes from their (...) primitives, and so I will turn to scrutinize the notion of "power" and "explanatory". (shrink)

This chapter contains sections titled: 1 2 3 Acknowledgment Bibliography.

Assuming for the sake of discussion that there is an external world, The "core" thesis of scientific realism is that some of our empirical beliefs (including the so-Called theoretical beliefs) succeed in correctly describing, In some important measure, The external world. Classical scientific realism also asserts that we are able to say justifiably just "which" of our beliefs so succeed in correctly describing the external world. This paper does not examine this last claim. Rather it seeks to defend the core (...) thesis of scientific realism and will assume (for reasons of limited space) the existence of an external world. In defending the core thesis, This paper returns to and defends the explanationist defense which asserts that the core thesis offers the best available explanation for the long-Term predictive success of "some" empirical or scientific hypotheses. The defense consists in examining and rejecting all available alternative explanations as well as all objections to the explanationist defense as offered in this paper. (shrink)

In earlier work ( Cleland [2001] , [2002]), I sketched an account of the structure and justification of ‘prototypical’ historical natural science that distinguishes it from ‘classical’ experimental science. This article expands upon this work, focusing upon the close connection between explanation and justification in the historical natural sciences. I argue that confirmation and disconfirmation in these fields depends primarily upon the explanatory (versus predictive or retrodictive) success or failure of hypotheses vis-à-vis empirical evidence. The account of historical explanation that (...) I develop is a version of common cause explanation. Common cause explanation has long been vindicated by appealing to the principle of the common cause. Many philosophers of science (e.g., Sober and Tucker) find this principle problematic, however, because they believe that it is either purely methodological or strictly metaphysical. I defend a third possibility: the principle of the common cause derives its justification from a physically pervasive time asymmetry of causation (a.k.a. the asymmetry of overdetermination). I argue that explicating the principle of the common cause in terms of the asymmetry of overdetermination illuminates some otherwise puzzling features of the practices of historical natural scientists. (shrink)

This paper examines the status of explanation in the natural sciences and ethics by focusing on the important role of empirical evidence and theoretical properties. As a means of exploring these issues, the debate between Nicholas Sturgeon and Gilbert Harman will serve as a central point in the discussion, since Sturgeon has provided several arguments against Harman's attempt to draw a distinction between scientific and moral explanation. Specifically, Sturgeon holds that the special function of observation and testing, which we commonly (...) believe to be relevant only in science, can be applied to moral cases with equal justification. Overall, Sturgeon's arguments will be exposed as deficient in several respects, which is an aspect of this debate often overlooked. (shrink)

In an explanation, what does the explaining and what gets explained? What are the relata of the explanation relation? Candidates include: people, events, facts, sentences, statements, and propositions.

Many people argue that history makes a special difference to the subjects of biology and psychology, and that history does not make this special difference to other parts of the world. This paper will show that historical properties make no more or less of a difference to biology or psychology than to chemistry, physics, or other sciences. Although historical properties indeed make a certain kind of difference to biology and psychology, this paper will show that historical properties make the same (...) kind of difference to geology, sociology, astronomy, and other sciences. Similarly, many people argue that nonhistorical properties make a special difference to the nonbiological and the nonpsychological world. This paper will show that nonhistorical properties make the same difference to all things in the world when it comes to their causal behavior and that historical properties make the same difference to all things in the world when it comes to their distributions. Although history is special, it is special in the same way to all parts of the world. (shrink)

Although explanation is widely regarded as an important concept in the study of rational inquiry, it remains largely unexplored outside the philosophy of science. This, I believe, is not due to oversight as much as to institutional resistance. In analytic philosophy it is basic that epistemic rationality is a function of justification and that justification is a function of argument. Explanation, however, is not argument nor is belief justification its function. I argue here that the task of incorporating explanation into (...) the theory of rational inquiry poses a serious challenge to our basic concept of epistemic rationality as well as the a priori method of inquiry that still lies at the heart of analytic philosophy. Specifically, it pushes us toward a much stronger form of naturalism than is generally thought necessary, one in which argument and explanation are recognized as distinct and equally fundamental cognitive processes whose dynamic relationship is one of the central issues in the theory of rationality. (shrink)

Moral philosophers are, among other things, in the business of constructing moral theories. And moral theories are, among other things, supposed to explain moral phenomena. Consequently, one’s views about the nature of moral explanation will influence the kinds of moral theories one is willing to countenance. Many moral philosophers are (explicitly or implicitly) committed to a deductive model of explanation. As I see it, this commitment lies at the heart of the current debate between moral particularists and moral generalists. In (...) this paper I argue that we have good reasons to give up this commitment. In fact, I show that an examination of the literature on scientific explanation reveals that we are used to, and comfortable with, non-deductive explanations in almost all areas of inquiry. As a result, I argue that we have reason to believe that moral explanations need not be grounded in exceptionless moral principles. (shrink)

. This paper proposes a generalization of the theory of the process of explanation to include consistent as well as inconsistent situations. The generalization is strong, for example in the sense that, if the background theory and the initial conditions are consistent, it leads to precisely the same results as the theory from the lead paper (Halonen and Hintikka 2004). The paper presupposes (and refers to arguments for the view that) inconsistencies constitute problems and that scientists try to resolve them.

This paper is an empirical critique of causal accounts of scientific explanation. Drawing on explanations which rely on nonlinear dynamical modeling, I argue that the requirement of causal relevance is both too strong and too weak to be constitutive of scientific explanation. In addition, causal accounts obscure how the process of mathematical modeling produces explanatory information. I advance three arguments for the inadequacy of causal accounts. First, I argue that explanatorily relevant information is not always information about causes, even in (...) cases where the explanandum has an identifiable causal history. Second, I argue that treating theoretical explanations as reductions from general causal laws does not accurately describe the types of "top-down" explanations typical of dynamical modeling. Finally, I argue that causal/mechanical accounts of explanation are intrinsically vulnerable to the irrelevance problem. (shrink)

Selection explanations explain some non-accidental generalizations in virtue of a selection process. Such explanations are not particulaizable - they do not transfer as explanations of the instances of such generalizations. This is unlike many explanations in the physical sciences, where the explanation of the general fact also provides an explanation of its instances (i.e. standard D-N explanations). Are selection explanations (e.g. in biology) therefore a different kind of explanation? I argue that to understand this issue, we need to see that (...) a standard D-N explanation of some non-accidental generalization (al Fs are Gs) may also ipso facto explain its contrapositive (all non-Gs are non-Fs), but the explanation is particularizable with respect to the former but not to the latter. This can be seen by noting that the Raven Paradox counterexample to the H-D model of confirmation also generates a counterexample to the D-N model of explanation (all ravens are black does not explain why the non-black shoe is a non-raven). In such cases it is natural to take the generalization with the positive predicates to have a particularizable explanation. However, this need not be the case, and in selection explanations it is the generalization with the positive predicates whose explanation is no particularizable. Thus there is no need to suppose that selection explanations are fundamentally different. (shrink)

There has been a recent focused effort in philosophical scholarship to bridge the perceived divide between pragmatism and analytic philosophy. This divide, it has been suggested, is over philosophical doctrines, methods, and even aims. This is not to say there has not been fruitful—even if antagonistic—dialogue between these two philosophical traditions. Clearly there has been, e.g., Russell's famous (or infamous) disputes with James and Dewey. Clearly also, there has been direct philosophical influence from one tradition to the other, e.g., Peirce (...) and Dewey on Quine's naturalistic works. Nevertheless, all too often the seminal works of pragmatists, especially the classical "Big Three," have not been mined by analytic philosophers, e.g., Peirce's writings on vagueness or on reference. This neglect has been, I believe, to the detriment of analytic philosophy. I intend this paper to be one step in correcting this neglect by suggesting both that Peirce's take on (scientific) explanation is a corrective to the prevailing models of explanation and that these models supply (at least in part) fecund additions to Peirce's views. First I will lay out four contemporary models of... (shrink)