This paper examines the explanatory gap account. The key notions for its proper understanding are analysed. In particular, the analysis is concerned with the role of “thick” and “thin” modes of presentation and “thick” and “thin” concepts which are relevant for the notions of “thick” and “thin” conceivability, and to that effect relevant for the gappy and non-gappy identities. The last section of the paper discusses the issue of the intelligibility of explanations. One of the conclusions is that the (...) explanatory gap account only succeeds in establishing the epistemic gap. The claim that psychophysical identity is not intelligibly explicable, and thus opens the explanatory gap, would require an indepen- dent argument which would prove that intelligible explanations stem only from conceptual analysis. This, I argue, is not the case. (shrink)

Explanation-Aware Computing: Papers from the 2007 AAAI Workshop, Association for the Advancement of Artificial Intelligence, Technical Report WS-07-06, Menlo Park California, AAAI Press, 2007, 1-9.

The question of liability in the case of using intelligent agents is far from simple, and cannot sufficiently be answered by deeming the human user as being automatically responsible for all actions and mistakes of his agent. Therefore, this paper is specifically concerned with the significant difficulties which might arise in this regard especially if the technology behind software agents evolves, or is commonly used on a larger scale. Furthermore, this paper contemplates whether or not it is possible to share (...) the responsibility with these agents and what are the main objections surrounding the assumption of considering such agents as responsible entities. This paper, however, is not intended to provide the final answer to all questions and challenges in this regard, but to identify the main components, and provide some perspectives on how to deal with such issue. (shrink)

This paper offers a dialogue theory of explanation. A successful explanation is defined as a transfer of understanding in a dialogue system in which a questioner and a respondent take part. The questioner asks a special sort of why-question that asks for understanding of something and the respondent provides a reply that transfers understanding to the questioner. The theory is drawn from recent work on explanation in artificial intelligence (AI), especially in expert systems, but applies to scientific, (...) legal and everyday conversational explanations. (shrink)

Teleological explanations (TEs) account for the existence or properties of an entity in terms of a function: we have hearts because they pump blood, and telephones for communication. While many teleological explanations seem appropriate, others are clearly not warranted-for example, that rain exists for plants to grow. Five experiments explore the theoretical commitments that underlie teleological explanations. With the analysis of [Wright, L. (1976). Teleological Explanations. Berkeley, CA: University of California Press] from philosophy as a point of departure, we examine (...) in Experiment 1 whether teleological explanations are interpreted causally, and confirm that TEs are only accepted when the function invoked in the explanation played a causal role in bringing about what is being explained. However, we also find that playing a causal role is not sufficient for all participants to accept TEs. Experiment 2 shows that this is not because participants fail to appreciate the causal structure of the scenarios used as stimuli. In Experiments 3-5 we show that the additional requirement for TE acceptance is that the process by which the function played a causal role must be general in the sense of conforming to a predictable pattern. These findings motivate a proposal, Explanation for Export, which suggests that a psychological function of explanation is to highlight information likely to subserve future prediction and intervention. We relate our proposal to normative accounts of explanation from philosophy of science, as well as to claims from psychology and artificial intelligence. (shrink)

This essay argues that, despite the failure of demarcation criteria for separating science from non-science, the mathematics of design and design-theoretic inferences nonetheless satisfy all the criteria of various competing theories of scientific explanation.

The intelligibility of nature was a persistent theme of William A. Wallace, OP, one of the most prolific Catholic scholars of the late twentieth century. This Reader aims to make available a representative selection of his work in the history of science, natural philosophy, and theology illustrating his defense and development of this central theme. Wallace is among the most important Galileo scholars of the past fifty years and a key figure in the recent revival of scientific realism. Further, (...) his long and productive scholarly career has been shaped by a continuous effort to bring the resources of the Aristotelian-Thomistic tradition to the solution of contemporary problems of philosophy and science. Through all of these contributions, Wallace has provided the foundation for a renewed confidence in the capacity of human knowers to attain understanding of the natural order. Consequently, the overall aim of this volume is to secure continued access to his scholarship for readers in the new millennium. -/- Intelligibility of Nature contains twenty-nine previously published essays written by Wallace over a period of some forty years. Many of these essays are currently not readily accessible. They are arranged in five thematic groups, each representing a major subject-area of Wallace's scholarly interests. The first group is devoted to essays on making nature intelligible through the use of scientific models. The second group of essays investigates various ways in which the Aristotelian-Thomistic tradition is foundational to contemporary scientific research. Essays in the third group are historical studies on the origins of modern science. The fourth group of essays discuss the viability of the cosmological argument for the existence of God in light of natural science. The final group of essays consider the relation of science and religion. Together these essays provide a representative sample of Wallace's multifaceted contributions to scholarship. (shrink)

Explanation-Aware Computing: Papers from the 2007 AAAI Workshop, Association for the Advancement of Artificial Intelligence, Technical Report WS-07-06, Menlo Park California, AAAI Press, 2007, 1-9.

In trying to make sense out of our behavior, we reach a point at which we stop talking about what we did and start talking about what we wish we had done, about what we mean to do next. But we think we are still talking about our motives and intentions in what we did. How do we know when we cross the line between finding out what actually happened and ascribing to a situation what we think ought to have (...) happened?This is a serious question, precisely because we usually do not know the difference between what we do and what we imagine that we do. For since all action is governed by our conception of “how we would have it” and “what we would bring about,” we can always honestly claim to have meant what we did for the best. We frequently find ourselves in the situation in which we don't know why we did what we did, but by casting about, we can always turn up a purpose for which our action must have been meant - a purpose which we now perceive.The reasons we ascribe to our actions, like our excuses, are governed by the demands made on us now, rather than by the factual content of what we did. Thus the behavior of reason-giving or excusing is meant to satisfy the social situation in which we presently find ourselves: the demand for explanation, the need to be “in the right,” the need to reinforce our convictions. The verbal account of what we did is thus literally about (as representing and serving) where we are now rather than what we did then.Our verbal account is about the context of social obligation in which we have to explain ourselves, not about the behavior we claim to interpret. This is why, proverbially, we will give any excuse which will satisfy the accuser and why “understanding” i.e. inventing reasons for our conduct, is no corrective. The addict, alcoholic, neurotic and criminal recidivist are frequently rational and even persuasive in giving reasons for what they do. And we wonder why if they understand their problems so well they can't do something about them. The problem is that the habitué doesn't understand his behavior at all. He is simply trained to give a certain after the fact explanation as a conditioned response to his habitual behavior. What he “understands” is not what makes him go back to the pill, the bottle or to crime, but what to tell people each time after he does. Naturally, when the habit-triggering occasion arises, it is never that reason - the stated one - which leads to repetition, but some new, special crisis, difficulty or impulse.Bluntly stated, our reasons and justifications have little or nothing to do with our actual behavior. We continue to suppose that we intend to do this or that immediately before we do it. But insofar as this intention is expressed it is a separate verbal act having its own conditions and consequences; insofar as it is unexpressed (and in general) intention is disclosed only through conduct. It is only the powerful social convention of justification that supports the illusion that verbal expressions are causal antecedents of acts.Two conclusions follow. First, the “sense” of our activity is either evident in the activity or else it is nonexistent, since not yet discovered meaning is meaning not yet achieved. Second, our verbal justifications are acts having a peculiar sense of their own. They express the relation of our convictions and objectives to the convictions and objectives of those to whom our explanations are addressed. Here is where, in the process of making sense of our activity, the crossover from discovery to invention occurs. This is why we may never discover the extent to which our justifications are inventions superimposed on the conduct they are supposed to represent. Insofar as our justification satisfies those to whom it is addressed, (including ourselves) it is right. But we take this rightness, which is actually the the compatibility of our values and objectives, to be the rightness of the act in question. The reasons we give for our actions make great sense, but despite our invincible “rightness” our actions, taken in themselves, make no more sense.What are our options? They are either to put ourselves in the power of those to whom we explain ourselves, or to take responsibility for our own nonsense. Take our conduct out of the medium of talk - of habitual justification - in which it is suspended and sustained, and what is it? Whatever has become of the “importance” of our appointments, our meetings, our appearances, our obligations, our consistency? Experience, not persuasion and reasoning, is the corrective. Only the direct experience of futility is powerful enough to break our addiction to justification. (shrink)

Early modern experimental philosophers were opposed to speculation, and yet many endorsed speculative theories. This chapter gives a partial explanation of why this is so, using Robert Boyle’s acceptance and promotion of the corpuscular philosophy as a case study. It argues that, in addition to furnishing experimental evidence for the corpuscular hypothesis in his Forms and Qualities, Boyle attempted to establish its epistemic superiority over other speculative theories on the grounds that it is founded upon superior principles. In his (...) ‘Excellency and Grounds of the Mechanical Hypothesis’, Boyle claims that the principles of the corpuscular philosophy are more intelligible, more parsimonious, ontologically primary, simpler and more comprehensive than those of its rivals. On grounds of comparative plausibility then, according to Boyle, the corpuscular philosophy is eminently preferable to its rivals. The final section of the chapter traces the legacy of Boyle’s appraisal of the epistemic status of the corpuscular philosophy in the writings of many of his compatriots from the mid-1660s to the Cyclopedia of Ephraim Chambers of 1728. (shrink)

Existential phenomenologists hold that the two most basic forms of intelligent behavior, learning, and skillful action, can be described and explained without recourse to mind or brain representations. This claim is expressed in two central notions in Merleau-Ponty's Phenomenology of Perception: the intentional arc and the tendency to achieve a maximal grip. The intentional arc names the tight connection between body and world, such that, as the active body acquires skills, those skills are stored, not as representations in the mind, (...) but as dispositions to respond to the solicitations of situations in the world. A phenomenology of skill acquisition confirms that, as one acquires expertise, the acquired know-how is experienced as finer and finer discriminations of situations paired with the appropriate response to each. Maximal grip names the body's tendency to refine its responses so as to bring the current situation closer to an optimal gestalt. Thus, successful learning and action do not require propositional mental representations. They do not require semantically interpretable brain representations either.Simulated neural networks exhibit crucial structural features of the intentional arc, and Walter Freeman's account of the brain dynamics underlying perception and action is structurally isomorphic with Merleau-Ponty's account of the way a skilled agent is led by the situation to move towards obtaining a maximal grip. (shrink)

Explanation and understanding are intimately connected notions, but the nature of that connection has generally not been considered a topic worthy of serious philosophical investigation. Most authors have avoided making reference to the notion of understanding in their accounts of explanation because they fear that any mention of the epistemic states of the individuals involved compromises the objectivity of explanation. Understanding is a pragmatic notion, they argue, and pragmatics should be kept at a safe distance from the (...) universal features of explanation. My main contention in this dissertation is that there is a sense in which it is meaningful and useful to talk about objective understanding, and that to characterize this notion it is necessary to formulate an account of objective explanation that makes reference to the beliefs and epistemic goals of the participants in a cognitive enterprise. ;My account of explanation is based on the belief-doubt model of inquiry first proposed by Peirce. Using the tools provided by decision theory, and the models of belief revision developed by Isaac Levi and by Alchourron, Gardenfors, and Makinson, I analyze the objective and pragmatic conditions that a piece of information must fulfill to be both explanatory and epistemically useful to an inquiring agent. The objective basis for an explanation is provided by the relation of probabilistic relevance that the fact described by the explanans bares to the fact described by the explanandum. But an explanation is much more than its foundation. As an epistemological notion, a potential explanation can only be a bona fide explanation if it becomes part of inquiry, that is, if an agent or a group of agents can see any value in it for their cognitive purposes. I provide a way to evaluate an explanation in terms of its credibility, its content, and its explanatory value. I argue that although an objective measure of the credibility and content of an explanation can be established, in many cases a complete determination of its explanatory value will ultimately depend on the interests and goals of the individual inquirers. (shrink)

Much of the philosophical discussion of explanations has centered around two broad conceptions of what sorts of ‘things’ explanations are – namely, the descriptive and ontic conceptions. Defenders of each argue that scientific psychology has at best little to contribute to the study of explanations. These anti-psychologistic arguments come in two main varieties, the metaphysical and the epistemic. Both varieties trace back to Hempel and recur in the more recent writings of prominent mechanists. The metaphysical arguments attempt to combat psychologism (...) about explanation by doubly dissociating explanations themselves from the commonly associated phenomenology of explanation. The epistemic arguments attempt to combat psychologism by doubly dissociating the good-making features of explanations from the CAPE. I focus on the metaphysical arguments here, for only they directly concern the nature of, as opposed to the evaluation conditions for, explanations. The metaphysical arguments are all shown to equivocate over ‘explanation’, and thus they are no threat to the proposal that sometimes explanations just are psychological events. Such arguments do, nonetheless, gesture towards something important – namely, that the CAPE is neither necessary nor sufficient for explanation in any sense. Similar considerations lead another latter-day mechanist to conclude that no phenomenology of any sort is essential for explanation. What all of these thinkers overlook, however, is that there is another important phenomenological correlate of explanation – namely, finding a happening intelligible or understanding how or why-possibly – and it may well be of the very essence of explanation, at least in the psychological sense. I propose a psychomechanistic theory of intelligibility that accounts for how we can, through finite means, make boundless, endlessly qualified inferences about the implications of our explanations. In closing I note that, although finding a happening intelligible does sometimes engender an unjustified confidence in correctness, we should not overlook how attaining this state does at least put us in a kind of epistemic base camp. I also hint at how intelligibility could turn out to be the ultimate source of the descriptive and ontic conceptions of explanation. (shrink)

We make how a person acts intelligible by revealing it as rational in the light of what she perceives, thinks, wants and so on. For example, we might explain that she reached out and picked up a glass because she was thirsty and saw that it contained water. In doing this, we are giving a causal explanation of her behaviour in terms of her antecedent beliefs, desires and other attitudes. Her wanting a drink and realizing that the glass contained (...) one caused her reaching out and grasping for it. This tells us how the action came about and makes sense of why it happened. At least, something broadly along these lines strikes me as a fairly crude and partial regimentation of our pretheoretic understanding of everyday action explanation. (shrink)

Commentators standardly ascribe to Spinoza a belief in an exceptionless conceptual closure of mental and physical realms: no intention can allow us to understand a bodily movement, no bodily injury can make intelligible a sensation of pain. This counterintuitive doctrine, most often now referred to as Spinoza's 'attribute barrier', has weighty repercussions for his views on intelligibility, nature of the mind, identity, and causality. I argue against the standard reading of the doctrine, by showing that it produces an inconsistent (...) epistemological picture, contradicting Spinoza's theory of mind and his commitment to universal intelligibility. The alternative account I propose is also philosophically more compelling and plausible as an account of thought in its essential intentionality. (shrink)

In this paper I critically examine the notion of explanation used in Artificial Intelligence in general, and in the theory of belief revision in particular. I focus on two of the best known accounts in the literature: Pagnucco’s abductive expansion functions and Gärdenfors’ counterfactual analysis. I argue that both accounts are at odds with the way in which this notion has historically been understood in philosophy. They are also at odds with the explanatory strategies used in actual scientific practice. (...) At the end of the paper I outline the desiderata for an epistemologically motivated, scientifically informed belief revision model for explanation. (shrink)

This paper presents a unified theory of reality, which integrates two interdependent frameworks: Cellular Primary Consciousness Theory (CPCT) and Open and Closed Systems Theory (OCST). Although CPCT and OCST can each stand as individual theories, they are, in this work, combined to form a cohesive explanation of both the mechanics and purpose of the universe. CPCT posits that consciousness is a fundamental aspect of all life, extending to even the simplest cells, rather than being an emergent property exclusive to (...) complex neural systems. Drawing upon evidence from quantum mechanics, cell biology, and systems theory, CPCT challenges traditional views on the origins of consciousness by suggesting that awareness is a universal property present from the cellular level. OCST, which forms the second component of this unified theory, examines the nature of the universe as a closed system subject to the laws of entropy. This theory explores how the universe is moving toward maximum entropy, where meaning and contrasts dissolve. However, beyond this closed system lies the realm of an open system—a domain of infinite potential that transcends the dualities and limitations of the physical universe. This open system offers a philosophical explanation for existence and the possibility of eternal meaning. Together, these two theories provide a novel and cohesive framework that addresses both the mechanics of consciousness and the ultimate purpose of the universe. The integration of CPCT and OCST bridges scientific evidence from quantum biology and systems theory with philosophical explorations of existence, presenting a comprehensive vision of reality as both a closed physical system and an open metaphysical one. This unified theory encourages a reexamination of our place in the cosmos, suggesting that beyond the observable universe lies an infinite realm of potential where consciousness and meaning endure. (shrink)

Providing meaningful and actionable explanations for end-users is a situated problem requiring the intersection of multiple disciplines to address social, operational, and technical challenges. However, the explainable artificial intelligence community has not commonly adopted or created tangible design tools that allow interdisciplinary work to develop reliable AI-powered solutions. This paper proposes a formative architecture that defines the explanation space from a user-inspired perspective. The architecture comprises five intertwined components to outline explanation requirements for a task: (1) the end-users’ (...) mental models, (2) the end-users’ cognitive process, (3) the user interface, (4) the Human-Explainer Agent, and (5) the agent process. We first define each component of the architecture. Then, we present the Abstracted Explanation Space, a modeling tool that aggregates the architecture’s components to support designers in systematically aligning explanations with end-users’ work practices, needs, and goals. It guides the specifications of what needs to be explained (content: end-users’ mental model), why this explanation is necessary (context: end-users’ cognitive process), to delimit how to explain it (format: Human-Explainer Agent and user interface), and when the explanations should be given. We then exemplify the tool’s use in an ongoing case study in the aircraft maintenance domain. Finally, we discuss possible contributions of the tool, known limitations or areas for improvement, and future work to be done. (shrink)

This paper outlines a procedure for assessing the quality of failure explanations in engineering failure analysis. The procedure structures the information contained in explanations such that it enables to find weak points, to compare competing explanations, and to provide redesign recommendations. These features make the procedure a good asset for critical reflection on some areas of the engineering practice of failure analysis and redesign. The procedure structures relevant information contained in an explanation by means of structural equations so as (...) to make the relations between key elements more salient. Once structured, the information is examined on its potential to track counterfactual dependencies by offering answers to relevant what-if-things-had-been-different questions. This criterion for explanatory goodness derives from the philosophy of science literature on scientific explanation. The procedure is illustrated by applying it to two case studies, one on Failure Analysis in Mechanical Engineering (a broken vehicle shaft) and one on Failure Analysis in Civil Engineering (a collapse in a convention center). The procedure offers failure analysts a practical tool for critical reflection on some areas of their practice while offering a deeper understanding of the workings of failure analysis (framing it as an explanatory practice). It, therefore, allows to improve certain aspects of the explanatory practices of failure analysis and redesign, but it also offers a theoretical perspective that can clarify important features of these practices. Given the programmatic nature of the procedure and its object (assessing and refining explanations), it extends work on the domain of computational argumentation. (shrink)

Increasingly secret, complex, and inscrutable computational systems are being used to intensify existing power relations and to create new ones; in particular, they are being used to govern. To be all-things-considered morally permissible new, or newly intense, power relations must meet standards of procedural legitimacy and proper authority. This is necessary for them to protect and realise democratic values of individual liberty, relational equality, and collective self-determination. For governing power in particular to be legitimate and have proper authority, it must (...) meet a publicity requirement: reasonably competent members of the governed community must be able to determine that they are being governed legitimately and with proper authority. The publicity requirement can be satisfied only if the powerful can explain their decision-making to members of their political community. At least some duties of explanation are therefore democratic duties. This chapter first sets out the foregoing argument, then applies it to opaque computational systems, and clarifies precisely what kinds of explanations are necessary to fulfil these democratic values. (shrink)

Many philosophers of science believe that empirical psychology can contribute little to the philosophical investigation of explanations. They take this to be shown by the fact that certain explanations fail to elicit any relevant psychological events (e.g., familiarity, insight, intelligibility, etc.). We report results from a study suggesting that, at least among those with extensive science training, a capacity to render an event intelligible is considered a requirement for explanation. We also investigate for whom explanations must be capable (...) of rendering events intelligible and whether or not accuracy is also viewed as a requirement. (shrink)

Among the many tensions in Leibniz’s philosophical system is his tendency to invoke both mechanistic and teleological explanations. Jonathan Bennett, typicalof recent Leibniz commentators, attempts to relieve this difficulty by arguing that teleology for Leibniz is theological posturing and philosophically thin; such a doctrine does not serve to explain the relationship between teleology and mechanism. I argue that Leibniz’s appeal to final causality is both inextricably grounded in his wider metaphysic and helpful in understanding the preconditions for causality in general. (...) To this end I defend the two following claims: 1) It is in part Leibniz’s theory of contingency, and not exclusively theological concerns, that leads him to conclude that the laws of nature must admit of teleological explanations. 2) The finality of the laws of nature, coupled with one of Leibniz’s most promising arguments against occasionalism, are together sufficient to show that teleology must play a role in explanations of the activity of all genuine substances. (shrink)

A key theme in the historiographical work of Machamer has been the ways that motion is made intelligible through explanatory means of natural motion and models of the simple machines such as the lever and pendulum. One way of spelling out the explanatory value of these strategies is through the concept of equilibrium. Natural motion and simple machines allow the simplification of complex problems in terms of self-evident, intelligible equilibrium conditions. This chapter connects the theme of equilibrium and natural motion (...) across Machamer’s work on mechanisms and mechanical explanation, on Aristotle, Galileo, Descartes and Newton, and on the pendulum and mental models. Just as equilibrium can be found within science, it also becomes a model of intelligibility for doing history and philosophy of science: a normative but objective representation of the important properties of science. (shrink)

This paper is concerned with a quality space model as an account of the intelligibility of explanation. I argue that descriptions of causal or functional roles (Chalmers Levine, 2001) are not the only basis for intelligible explanations. If we accept that phenomenal concepts refer directly, not via descriptions of causal or functional roles, then it is difficult to find role fillers for the described causal roles. This constitutes a vagueness constraint on the intelligibility of explanation. Thus, (...) I propose to use quality space models to develop a systematic way of studying different modalities of perception and feelings, e.g., visual and auditory perception, pain, and emotion, that can reveal some structural relations among these modalities. It might turn out that topological explanation can be more intelligible than causal explanation in this case. I discuss two accounts of a quality space for color vision (Clark, 2000; Rosenthal, 2010) and propose how to construct a quality space for pain. Daniel Kostic is Associated Researcher at Berlin School of Mind and Brain. (shrink)

In Aristotle's teleological view of the world, natural things come to be and are present for the sake of some function or end. Whereas much of recent scholarship has focused on uncovering the physical underpinnings of Aristotle's teleology and its contrasts with his notions of chance and necessity, this book examines Aristotle's use of the theory of natural teleology in producing explanations of natural phenomena. Close analyses of Aristotle's natural treatises and his Posterior Analytics show what methods are used for (...) the discovery of functions or ends that figure in teleological explanations, how these explanations are structured, and how well they work in making sense of phenomena. The book will be valuable for all who are interested in Aristotle's natural science, his philosophy of science, and his biology. (shrink)

Some contemporary theories about the origin and the nature of language resort to concepts with no bearing on Darwinian evolutionary hypothesis or evo-devo perspective which are both based on the reconstruction of species morphological structure transformation. These theories, which evoke qualitative leap, cultural evolution, structure/function coevolution as esplicative principles for human evolution, in our opinion, result compatible in some points with the most recent Intelligent Design accounts. Attempting to substantiate itself as a scientific theory, the contemporary ID is ready to (...) give up creationist explanation just to impeding Darwin’s fundamental idea according to which it’s possible to explain evolution only through a gradual material modification of structures. For comprehending a complex phenomenon as human language – according to ID – it’s necessary appealing to a second substance, whatever it is. This idea seems to be at the bottom of all those theories which have rejected monistic structural explanation for embracing functional, psychological or cultural accounts. We consider these kinds of explanation real unresolved residuals of ID, residuals nested in the heart of the most accredited scientific theories. (shrink)

Artificial intelligence (AI) and robotics are digital technologies that will have significant impact on the development of humanity in the near future. They have raised fundamental questions about what we should do with these systems, what the systems themselves should do, what risks they involve, and how we can control these. - After the Introduction to the field (§1), the main themes (§2) of this article are: Ethical issues that arise with AI systems as objects, i.e., tools made and used (...) by humans. This includes issues of privacy (§2.1) and manipulation (§2.2), opacity (§2.3) and bias (§2.4), human-robot interaction (§2.5), employment (§2.6), and the effects of autonomy (§2.7). Then AI systems as subjects, i.e., ethics for the AI systems themselves in machine ethics (§2.8) and artificial moral agency (§2.9). Finally, the problem of a possible future AI superintelligence leading to a “singularity” (§2.10). We close with a remark on the vision of AI (§3). - For each section within these themes, we provide a general explanation of the ethical issues, outline existing positions and arguments, then analyse how these play out with current technologies and finally, what policy consequences may be drawn. (shrink)

The article addresses the problem of identifying methods to develop the ability of artificial intelligence (AI) systems to provide explanations for their findings. This issue is not new, but, nowadays, the increasing complexity of AI systems is forcing scientists to intensify research in this direction. Modern neural networks contain hundreds of layers of neurons. The number of parameters of these networks reaches trillions, genetic algorithms generate thousands of generations of solutions, and the semantics of AI models become more complicated, going (...) to the quantum and non-local levels. The world’s leading companies are investing heavily in creating explainable AI (XAI). However, the result is still unsatisfactory: a person often cannot understand the “explanations” of AI because the latter makes decisions differently than a person, and perhaps because a good explanation is impossible within the framework of the classical AI paradigm. AI faced a similar problem 40 years ago when expert systems contained only a few hundred logical production rules. The problem was then solved by complicating the logic and building added knowledge bases to explain the conclusions given by AI. At present, other approaches are needed, primarily those that consider the external environment and the subjectivity of AI systems. This work focuses on solving this problem by immersing AI models in the social and economic environment, building ontologies of this environment, taking into account a user profile and creating conditions for purposeful convergence of AI solutions and conclusions to user-friendly goals. (shrink)

A collection of fourteen essays which provide an overview of the argument for intelligent design, with diagrams, explanations, and relevant quotations.

Epidemiological explanation often has a “black box” character, meaning the intermediate steps between cause and effect are unknown. Filling in black boxes is thought to improve causal inferences by making them intelligible. I argue that adding information about intermediate causes to a black box explanation is an unreliable guide to pragmatic intelligibility because it may mislead us about the stability of a cause. I diagnose a problem that I call wishful intelligibility, which occurs when scientists misjudge (...) the limitations of certain features of an explanation. Wishful intelligibility gives us a new reason to prefer black box explanations in some contexts. (shrink)

Explainable artificial intelligence (XAI) aims to help people understand black box algorithms, particularly of their outputs. But what are these explanations and when is one explanation better than another? The manipulationist definition of explanation from the philosophy of science offers good answers to these questions, holding that an explanation consists of a generalization that shows what happens in counterfactual cases. Furthermore, when it comes to explanatory depth this account holds that a generalization that has more abstract variables, (...) is broader in scope and/or more accurate is better. By applying these definitions and contrasting them with alternative definitions in the XAI literature I hope to help clarify what a good explanation is for AI. (shrink)

In books such as The World Within the World and The Anthropic Cosmological Principle, astronomer John Barrow has emerged as a leading writer on our efforts to understand the universe. Timothy Ferris, writing in The Times Literary Supplement of London, described him as "a temperate and accomplished humanist, scientist, and philosopher of science--a man out to make a contribution, not a show." Now Barrow offers the general reader another fascinating look at modern physics, as he explores the quest for a (...) single, unifying theory that will unlock nature's secrets. Theories of Everything is more than a history of science, more than a popular report on recent research and discoveries. Barrow provides a reflective, intelligent commentary on what a true Theory of Everything would be--its ingredients, its limitations, and what it could tell us about the universe. Never before, he writes, have physicists been so confident and so eager in the hunt for this "cosmic Rosetta Stone," as he calls it: "a single all-embracing picture of all the laws of nature from which the inevitability of all things seen must follow with unimpeachable logic." He lays out eight essential ingredients for a Theory of Everything and then explores each in turn, tracing how our knowledge has developed and how scientific discovery relates to our changing philosophy and religious thought in each area. Some of these ingredients are obvious--the laws of nature must be explained, for example, as well as its organizing principles--but others may be surprising, such as broken symmetries and selection biases. A Theory of Everything must account for the fact that the universe is "messy and complicated," he tells us, and for the limitations imposed by the questions we ask and the information we can obtain. The key lies in the remarkable capacity of mathematics to express the fundamental workings of the physical world--a language that the human mind is uniquely equipped to understand and manipulate. Barrow examines what mathematics actually is and describes how it makes the universe intelligible and provides a path to the underlying coherence in nature--which has led, in fact, to arguments that the universe itself is a vast computer. Yet even the most complete theory, even the most comprehensive mathematical explanation, cannot account for the uncomputable varieties of human experience and thought. "No non-poetic account of reality," he writes, "can be complete." In a field where the authorities converse in equations and mathematical notations, John Barrow speaks with the voice of thoughtful and knowledgeable humanist. Written with eloquence and expertise, Theories of Everything establishes a new perspective on humanity's efforts to explain the universe. (shrink)

Some of human evil is a function of the historical stage of society. The evils and wickednesses of bureaucracy are as old as well‐developed bureaucratic hierarchies. Evil‐doers have character traits that may form recognizable patterns with explanatory weight. Evil‐doers produce reasons for their evil‐doing and offer justifications for their evil deeds. Psychological experiments may indeed establish important correlations and statistical probabilities that may be crucial for the formation of intelligent social policy. The greatest students of the place of evil in (...) human life and of the subjective psychology of evil in a complex social setting are great novelists, dramatists, and poets. From the late eighteenth century onwards ideological beliefs played a major role in stimulating evil, not for the greater glory of God but for the future glory of man. The first major social movement that bent an ideology to wholly evil policies were the Jacobins during the French Revolution. (shrink)

In his recent anthology, Intelligent Design Creationism and Its Critics, Robert Pennock continues his attack on what he considers to be the pseudoscience of Intelligent Design Theory. In this critical review, I discuss the main issues in the debate. Although the rhetoric is often heavy and the articles are intentionally stacked against Intelligent Design, there are many interesting topics in the philosophy of science to be found. I conclude that, contra Pennock, there is nothing intrinsically unscientific about Intelligent Design. At (...) this stage, however, it remains more of a provocative idea than a research program. Whether design theorists can bridge this gap is still very much in question. In any case, the debate serves as a modern case study for such classic problems as the nature of scientific explanations, theory change, the demarcation problem, and the role of metaphysical assumptions in the development of science. (shrink)

Merleau-ponty's thesis of the ambiguity of time is used to criticize husserl's and hume's views of the self. the thesis states that 1) the experience of present and past time depend on one another for intelligibility, and 2) the "objective" is past experience frozen in time; the "subjective" is present experience. adequate accounts of the self (and, generally, adequate philosophical explanation) will respect the ambiguity of time and discuss the interaction between subjective and objective facets of experience.

When it comes to the question of what kind of moral claim an intelligent or autonomous machine might have, one way to answer this is by way of comparison with humans: Is there a fundamental difference between humans and other entities? If so, on what basis, and what are the implications for science and ethics? This question is inherently imprecise, however, because it presupposes that we can readily determine what it means for two types of entities to be sufficiently different—what (...) I will refer to as being “discontinuous”. In this paper, I will sketch a formal characterization of what it means for types of entities to be unique with regard to each other. This expands upon Bruce Mazlish’s initial formulation of what he terms a continuity between humans and machines, Alan Turing’s epistemological approach to the question of machine intelligence, and Sigmund Freud’s notion of scientific revolutions dealing blows to the self-esteem of mankind. I will discuss on what basis we should regard entities as (dis-)continuous, the corresponding moral and scientific implications, as well as an important difference between what I term downgrading and upgrading continuities—a dramatic difference in how two previously discontinuous types of entities might become continuous. All of this will be phrased in terms of which scientific levels of explanation we need to presuppose, in principle or in practice, when we seek to explain a given type of entity. The ultimate purpose is to provide a framework that defines which questions we need to ask if we argue that two types of entities ought (not) to be explained (hence treated) in the same manner, as well as what it takes to reconsider scientific and ethical hierarchies imposed on the natural and artificial world. (shrink)

Since approval of the EU General Data Protection Regulation (GDPR) in 2016, it has been widely and repeatedly claimed that the GDPR will legally mandate a ‘right to explanation’ of all decisions made by automated or artificially intelligent algorithmic systems. This right to explanation is viewed as an ideal mechanism to enhance the accountability and transparency of automated decision-making. However, there are several reasons to doubt both the legal existence and the feasibility of such a right. In contrast (...) to the right to explanation of specific automated decisions claimed elsewhere, the GDPR only mandates that data subjects receive meaningful, but properly limited, information (Articles 13-15) about the logic involved, as well as the significance and the envisaged consequences of automated decision-making systems, what we term a ‘right to be informed’. Further, the ambiguity and limited scope of the ‘right not to be subject to automated decision-making’ contained in Article 22 (from which the alleged ‘right to explanation’ stems) raises questions over the protection actually afforded to data subjects. These problems show that the GDPR lacks precise language as well as explicit and well-defined rights and safeguards against automated decision-making, and therefore runs the risk of being toothless. We propose a number of legislative and policy steps that, if taken, may improve the transparency and accountability of automated decision-making when the GDPR comes into force in 2018. (shrink)

This essay develops a general account of one type of explanation found in history in particular: that an individual action is conceived as an exemplification of a rather complex schema of practical inference, under the provision that the facts which instantiate the various terms of the schema have an intelligible connection to one another. The essay then raises the question whether historians, anthropologists, and their contemporaneous audience can have an internal understanding of the actions of others, where those others (...) come from radically different cultures or times from the historians or anthropologists. An account is offered that, arguably, can resolve this problem and do justice to both the claim of internal understanding and the presumed cultural differentness between the agents studied and the historians and anthropologists who do the study. (shrink)

In exploring the nature of psychological explanation, this book looks at how psychologists theorize about the human ability to calculate, to speak a language and the like. It shows how good theorizing explains or tries to explain such abilities as perception and cognition. It recasts the familiar explanations of "intelligence" and "cognitive capacity" as put forward by philosophers such as Fodor, Dennett, and others in terms of a theory of explanation that makes established doctrine more intelligible to professionals (...) and their students.In particular, the book shows that vestigial adherence to the positivists' D-N model has distorted the view of philosophers of science about what psychologists (and biologists) do and has masked the real nature of explanation. Major sections in the book cover Analysis and Subsumption; Functional Analysis; Understanding Cognitive Capacities; and Historical Reflections.Robert Cummins is Associate Professor of Philosophy at the University of Illinois, Chicago Circle. A Bradford Book. (shrink)

This article reveals that explanatory practice in software engineering is in accordance with pragmatic explanatory pluralism, which states that explanations should at least partially be evaluated by their practical use. More specifically, I offer a defense of the idea that several explanation-types are legitimate in software engineering, and that the appropriateness of an explanation-type depends on (a) the engineer’s interests, and (b) the format of the explanation-seeking question he asks, with this format depending on his interests. This (...) idea is defended by considering examples that are representative for explanatory practice in software engineering. Different kinds of technological explanation are spelled out, and the dependence of their appropriateness on interests and question-formats is extensively illustrated. (shrink)

In Artificial Intelligence, Artificial Neural Networks are very accurate models in tasks such as classification and regression in the study of natural phenomena, but they are considered “black boxes” because they do not allow direct explanation of what they address. This paper reviews the possibility of scientific explanation from these models and concludes that other efforts are required to understand their inner workings. This poses challenges to access scientific explanation through their use, since the nature of Artificial (...) Neural Networks makes it difficult at first instance the scientific understanding that can be extracted from them. (shrink)

In lieu of an abstract, here is a brief excerpt of the content:94. HUME'S EXPLANATION OF RELIGIOUS BELIEF1 In The Natural History of Religion, David Hume offers a not unsophisticated account of the fact that persons hold religious beliefs. In so doing, he produces an explanatory system analogous to that which occurs concerning causal belief, belief in 'external objects', and belief in an enduring self in the Treatise ¦ The explanation of the occurrence of religious belief is more (...) detailed than the explanation provided in the other cases just mentioned. In the Natural History, Hume devotes a short volume to explaining religious belief, while in the Treatise the causal, external object, and enduring self beliefs merit 2 but long sections. More important, however, than length of treatment is the fact that the pattern of explanation is identical in each instance. The Natural History could be embedded without categorial clash into the Treatise, perhaps as Book Four with"fifteen sections, and each formerly separate volume would shed light on the program and tactics of the other. My interest here is in the epistemic features of the explanatory system Hume developed in the Natural History. Hume forthrightly proclaims that The Natural History Of Religion is in fact an attempt to explain the occurrence of religious belief. He writes: What those principles are, which give rise to the original belief, and what those accidents and causes are, which direct its operation, is the subject of our present enquiry.4 (NHR21) Original belief here does duty for "original religious belief", and as he takes religious belief to be nearly but not altogether universal in scope and astonishingly diverse in object, he supposes the principle, or cause, of such belief to be secondary in the sense that its operation is (so to say) defeasible and its product diversified. Hume's powerful critique of the argument from design in Sections II 95. through VIII of the Dialogues is not the only reason for doubting that his occasional kind remarks concerning it should be taken as indicating that he supposed it sound and valid. The very fact that Hume wrote a book intended to explain the occurrence of religious belief by identifying as its cause a built-in principle and its eliciting stimuli should give us pause about Hume's apparent acceptance of something like the argument from design. For while in the Natural History he says that: The whole frame of nature bespeaks an intelligent author; and no ¦ rational enquirer can, after serious reflection, suspend his belief a moment with regard to the primary principles of genuine Theism and Religion. (NHR21) He speaks, not of a conclusion having been proved true, but of a belief having been rendered unsuspendable. Further, neither we nor Hume will ordinarily offer a causal account of the fact that a person has a belief unless there is doubt that the person has sufficient reason for holding it. He does not, for example, offer any such explanation of our acceptance of sincere present-tense first person psychological reports, concerning the truth (indeed, the incorrigibility ) of which he in the Treatise confidently affirms : For since all actions and sensations of the mind are known to us by consciousness, they must necessarily appear in every particular what they are, and be what they appear. Every thing that enters the mind, being in rea lity perception, 'tis impossible any thing should to feeling appear different. This were to suppose that even where we are most intimately conscious, we might be mistaken. (T190) Or, even more modestly, and without assuming a causal-account and a sufficient-reason-account of a belief to be competing (or even necessarily different) explanations, we may note that Hume proposes to explain the occurrence of religious belief by reference to principle and eliciting stimuli without making reference to reasons or arguments as items possessing epistemic function or 96. evidential force, but only as items capable of triggering a built-in response. That this is_ his tactic I have argued rather fully on another occasion; here I will focus only on the general pattern of Hume's explanation of religious belief. Hume endeavors to explain religious belief without supposing it (in any of its forms) to be true as well as without supposing it... (shrink)