The paper provides a systematic overview of recent debates in epistemology and philosophy of science on the nature of understanding. We explain why philosophers have turned their attention to understanding and discuss conditions for “explanatory” understanding of why something is the case and for “objectual” understanding of a whole subject matter. The most debated conditions for these types of understanding roughly resemble the three traditional conditions for knowledge: truth, justification and belief. We discuss prominent views about how to construe these (...) conditions for understanding, whether understanding indeed requires conditions of all three types and whether additional conditions are needed. (shrink)

Epistemology has for a long time focused on the concept of knowledge and tried to answer questions such as whether knowledge is possible and how much of it there is. Often missing from this inquiry, however, is a discussion on the value of knowledge. In The Value of Knowledge and the Pursuit of Understanding Jonathan Kvanvig argues that epistemology properly conceived cannot ignore the question of the value of knowledge. He also questions one of the most fundamental assumptions in epistemology, (...) namely that knowledge is always more valuable than the value of its subparts. Taking Platos' Meno as a starting point of his discussion, Kvanvig tackles the different arguments about the value of knowledge and comes to the conclusion that knowledge is less valuable than generally assumed. Clearly written and well argued, this 2003 book will appeal to students and professionals in epistemology. (shrink)

Epistemic opacity vis a vis human agents has been presented as an essential, ineliminable characteristic of computer simulation models resulting from the characteristics of the human cognitive agent. This paper argues, on the contrary, that such epistemic opacity as does occur in computer simulations is not a consequence of human limitations but of a failure on the part of model developers to adopt good software engineering practice for managing human error and ensuring the software artefact is maintainable. One consequence of (...) such failures is to create a “technical debt” which manifests itself in the so-called novel confirmation holism confronted by complex systems models. The argument from the supposed essential epistemic opacity of computational science to a non-anthropocentric epistemology runs counter to best practice in software engineering and overlooks empirical results of software engineering science. (shrink)

Verification and validation are methods with which computer simulations are tested. While many practitioners draw a clear line between verification and validation and demand that the former precedes the latter, some philosophers have suggested that the distinction has been over-exaggerated. This chapter clarifies the relationship between verification and validation. Regarding the latter, validation of the conceptual and of the computational modelComputational model are distinguished. I argue that, as a method, verification is clearly different from validation of either of the models. (...) However, the methods are related to each other as follows: If we allow that the validation of the computational model need not include the comparisonComparison between simulation output and measured data, then the computational model may be validated by validating the conceptual model independently and by verifying the simulation with respect to it. This is often not realistic, however, because, in most cases, the conceptual modelConceptual model cannot be validated independently from the simulation. In such cases, the computational modelComputational model is verified with the aim to use it as an appropriate substitute for the conceptual model. Then simulation output is compared to measured data to validate both the computational and the conceptual model. I analyze the underlying inferences and argue that they require some prior confidence in the conceptual model and in verification. This suggests that verification precede validation that proceeds via a comparisonComparison between simulation outputOutput and measured data. Recent arguments to the effect that the distinction between verification and validation is not clear-cut do not refute these results, or so I argue against philosopher E. Winsberg. (shrink)

Where does the mind stop and the rest of the world begin? The question invites two standard replies. Some accept the demarcations of skin and skull, and say that what is outside the body is outside the mind. Others are impressed by arguments suggesting that the meaning of our words "just ain't in the head", and hold that this externalism about meaning carries over into an externalism about mind. We propose to pursue a third position. We advocate a very different (...) sort of externalism: an _active externalism_, based on the active role of the environment in driving cognitive processes. (shrink)

It is often claimed that scientists can obtain new knowledge about nature by running computer simulations. How is this possible? I answer this question by arguing that computer simulations are arguments. This view parallels Norton’s argument view about thought experiments. I show that computer simulations can be reconstructed as arguments that fully capture the epistemic power of the simulations. Assuming the extended mind hypothesis, I furthermore argue that running the computer simulation is to execute the reconstructing argument. I discuss some (...) objections and reject the view that computer simulations produce knowledge because they are experiments. I conclude by comparing thought experiments and computer simulations, assuming that both are arguments. (shrink)

What does it mean to understand something? I approach this question by comparing understanding with knowledge. Like knowledge, understanding comes, at least prima facia, in three varieties: propositional, interrogative and objectual. I argue that explanatory understanding (this being the most important form of interrogative understanding) and objectual understanding are not reducible to one another and are neither identical with, nor even a form of, the corresponding type of knowledge (nor any other type of knowledge). My discussion suggests that definitions of (...) the concepts of explanatory and of objectual understanding must include a commitment condition, a grasping condition, an answering-the-facts condition, and an epistemically internal justification condition, but no further external anti-luck condition. On this basis I argue against reducing explanatory understanding to propositional understanding, and in favour of identifying propositional understanding with propositional knowledge. (shrink)

The paper argues that an account of understanding should take the form of a Carnapian explication and acknowledge that understanding comes in degrees. An explication of objectual understanding is defended, which helps to make sense of the cognitive achievements and goals of science. The explication combines a necessary condition with three evaluative dimensions: an epistemic agent understands a subject matter by means of a theory only if the agent commits herself sufficiently to the theory of the subject matter, and to (...) the degree that the agent grasps the theory (i.e., is able to make use of it), the theory answers to the facts and the agent’s commitment to the theory is justified. The threshold for outright attributions of understanding is determined contextually. The explication has descriptive as well as normative facets and allows for the possibility of understanding by means of non-explanatory (e.g., purely classificatory) theories. (shrink)

Computer simulations are widely used in current scientific practice, as a tool to obtain information about various phenomena. Scientists accordingly rely on the outputs of computer simulations to make statements about the empirical world. In that sense, simulations seem to enable scientists to acquire empirical knowledge. The aim of this paper is to assess whether computer simulations actually allow for the production of empirical knowledge, and how. It provides an epistemological analysis of present-day empirical science, to which the traditional epistemological (...) categories cannot apply in any simple way. Our strategy consists in acknowledging the complexity of scientific practice, and trying to assess its rationality. Hence, while we are careful not to abstract away from the details of scientific practice, our approach is not strictly descriptive: our goal is to state in what conditions empirical science can rely on computer simulations. In order to do so, we need to adopt a renewed epistemological framework, whose categories would enable us to give a finer-grained, and better-fitted analysis of the rationality of scientific practice. (shrink)

Most computational analysts, as well as most governmental policy-makers and the public, view computational simulation accuracyAccuracy as a good agreementAgreement of simulation results with empirical measurements. However, decision-makers, such as business managers and safety regulators who rely on simulation for decision support, view computational simulation accuracy as much more than agreement of simulation results with experimental dataExperimental data. Decision-makers’ concept of accuracy is better captured by the term predictive capability of the simulation. Predictive capability meaning the use of a computational (...) modelComputational model to foretell or forecastBenchmark forecast the response of a system to conditions without availableExperimental data experimental dataData, even for system responses that have never occurred in nature. This chapter makes this important distinction by discussing the crucial ingredients needed for predictive capabilityPredictive capability: code verificationCode verification, solution verification, model validation, model calibrationCalibration, and predictive uncertaintyUncertainty estimation. Each of these ingredients is required, whether the simulation results are used in the generation of new knowledgeKnowledge, or for decision support by business managers, government policy-makers, or safety regulators. (shrink)

The practice of verification is grounded in mathematics highlighting the fundamental nature of its practice. Models of reality are fundamentally mathematical and verification assures the connection between the modeling intended and achieved in codeCode. Code verificationCode verification is a process where the correctness of a computer codeCode for simulation and modeling is proven. This “proof” is defined by the collection of evidence that the numerical approximationsApproximation are congruent with the model for the physical phenomena. The key metricMetric in codeCodeverificationCode verification (...) is the order of accuracyAccuracy of the approximationApproximation that should match theoretical expectations. In contrast, solution verification is an aspect of uncertaintyUncertainty estimationestimationEstimation associated with numerical errorError in simulations. Solution verification uses many of the same approaches as codeCodeverificationCode verification, but its principal outcome is an estimate of the numerical errorError. The order of convergenceConvergence is a secondary outcome. Together these two practices form an important part of the foundation of quality and credibilityCredibility in modeling and simulation. (shrink)

Simulation modeling, the core thesis of Calculated Surprises, is transforming the established conception of mathematical modeling in fundamental ways. These transformations feed back into philosophy of science, opening up new perspectives on longstanding oppositions. The book integrates historical features with both practical case studies and broad reflections on science and technology.

This article discusses agent-based simulation (ABS) as a tool of sociological understanding. I argue that agent-based simulations can play an important role in the expansion of explanatory understanding in the social sciences. The argument is based on an inferential account of understanding (Ylikoski 2009, Ylikoski & Kuorikoski 2010), according to which computer simulations increase our explanatory understanding by expanding our ability to make what-if inferences about social processes and by making these inferences more reliable. The inferential account also suggests a (...) number of ways in which the use of simulation methodology might give rise to illusory understanding. The structure of the article is as .. (shrink)

The ArgumentIn its reconstruction of scientific practice, philosophy of science has traditionally placed scientific theories in a central role, and has reduced the problem of mediating between theories and the world to formal considerations. Many applications of scientific theories, however, involve complex mathematical models whose constitutive equations are analytically unsolvable. The study of these applications often consists in developing representations of the underlying physics on a computer, and using the techniques of computer simulation in order to learn about the behavior (...) of these systems. In many instances, these computer simulations are not simple number-crunching techniques. They involve a complex chain of inferences that serve to transform theoretical structures into specific concrete knowledge of physical systems. In this paper I argue that this process of transformation has its own epistemology. I also argue that this kind of epistemology is unfamiliar to most philosophy of science, which has traditionally concerned itself with the justification of theories, not with their application. Finally, I urge that the nature of this epistemology suggests that the end results of some simulations do not bear a simple, straightforward relation to the theories from which they stem. (shrink)

That photography is a supremely realistic medium may be the commonsense view, but—as Edward Steichen reminds us—it is by no means universal. Dissenters note how unlike reality a photograph is and how unlikely we are to confuse the one with the other. They point to “distortions” engendered by the photographic process and to the control which the photographer exercises over the finished product, the opportunities he enjoys for interpretation and falsification. Many emphasize the expressive nature of the medium, observing that (...) photographs are inevitably colored by the photographer’s personal interests, attitudes, and prejudices.1 Whether any of these various considerations really does collide with photography’s claim of extraordinary realism depends, of course, on how that claim is to be understood.Those who find photographs especially realistic sometimes think of photography as a further advance in a direction which many picture makers have taken during the last several centuries, as a continuation or culmination of the post-Renaissance quest for realism.2 There is some truth in this. Such earlier advances toward realism include the development of perspective and modeling techniques, the portrayal of ordinary and incidental details, attention to the effects of light, and so on. From its very beginning, photography mastered perspective. Subtleties of shading, gradations of brightness nearly impossible to achieve with the brush, became commonplace. Photographs include as a matter of course the most mundane details of the scenes they portray—stray chickens, facial warts, clutters of dirty dishes. Photographic images easily can seem to be what painters striving for realism have always been after. 2. See André Bazin, “The Ontology of the Photographic Image,” What is Cinema?, trans. Hugh Gray, vol. 1, p. 12; all further references to this work, abbreviated “OPI,” will be included in the text. See also Rudolf Arnheim, “Melancholy Unshaped,” Toward a Psychology of Art: Collected Essays, p. 186. Kendall L. Walton is professor of Philosophy at the University of Michigan. He is currently completing a book on representation in the arts. (shrink)

This paper offers an analytic perspective on epistemic dependence that is grounded in theoretical discussion and field observation at the same time. When in the course of knowledge creation epistemic labor is divided, collaborating scientists come to depend upon one another epistemically. Since instances of epistemic dependence are multifarious in scientific practice, I propose to distinguish between two different forms of epistemic dependence, opaque and translucent epistemic dependence. A scientist is opaquely dependent upon a colleague if she does not possess (...) the expertise necessary to independently carry out, and to profoundly assess, the piece of scientific labor which her colleague is contributing. If the scientist does possess the necessary expertise, I argue, her dependence is translucent. However, the distinction between opaque and translucent epistemic dependence does not exhaust dependence relations in scientific practice, because many dependence relations are neither entirely opaque nor translucent. I will discuss why this is the case, and show how we can make sense of the gray zone between opaque and translucent epistemic dependence. (shrink)

We address some of the epistemological challenges highlighted by the Critical Data Studies literature by reference to some of the key debates in the philosophy of science concerning computational modeling and simulation. We provide a brief overview of these debates focusing particularly on what Paul Humphreys calls epistemic opacity. We argue that debates in Critical Data Studies and philosophy of science have neglected the problem of error management and error detection. This is an especially important feature of the epistemology of (...) Big Data. In “Error” section we explain the main characteristics of error detection and correction along with the relationship between error and path complexity in software. In this section we provide an overview of conventional statistical methods for error detection and review their limitations when faced with the high degree of conditionality inherent to modern software systems. (shrink)

It is argued that a popular way of accounting for the distinctive value of knowledge by appeal to the distinctive value of cognitive achievements fails because it is a mistake to identify knowledge with cognitive achievements. Nevertheless, it is claimed that understanding, properly conceived, is a type of cognitive achievement, and thus that the distinctive value of cognitive achievements can explain why understanding is of special value.

Today’s most sophisticated simulation studies of future climate employ not just one climate model but a number of models. I explain why this “ensemble” approach has been adopted—namely, as a means of taking account of uncertainty—and why a comprehensive investigation of uncertainty remains elusive. I then defend a middle ground between two camps in an ongoing debate over the transformation of ensemble results into probabilistic predictions of climate change, highlighting requirements that I refer to as ownership, justification, and robustness.

In 1904, Norwegian physicist Vilhelm Bjerknes published what would become a landmark paper in the history of meteorology. In that paper, he proposed that daily weather forecasts could be made by calculating later states of the atmosphere from an earlier state using the laws of hydrodynamics and thermodynamics (Bjerknes 1904). He outlined a set of differential equations to be solved and advocated the development of graphical and numerical solution methods, since analytic solution was out of the question. Using these theory-based (...) equations to produce daily forecasts, however, turned out to be more difficult than anticipated. Graphical solution techniques had limited success, and a first attempt to use .. (shrink)

This paper starts by looking at the coincidence of surprising behavior on the nanolevel in both matter and simulation. It uses this coincidence to argue that the simulation approach opens up a pragmatic mode of understanding oriented toward design rules and based on a new instrumental access to complex models. Calculations, and their variation by means of explorative numerical experimentation and visualization, can give a feeling for a model's behavior and the ability to control phenomena, even if the model itself (...) remains epistemically opaque. Thus, the investigation of simulation in nanoscience provides a good example of how science is adapting to a new instrument: computer simulation. (shrink)

Thought experiments and simulation experiments are compared and contrasted with each other. While the former rely on epistemic transparency as a working condition, in the latter complexity of model dynamics leads to epistemic opacity. The difference is elucidated by a discussion of the different kinds of iteration that are at work in both sorts of experiment.

My purpose in this essay is to clarify the notion of explanation by computer simulation in artificial intelligence and cognitive science. My contention is that computer simulation may be understood as providing two different kinds of explanation, which makes the notion of explanation by computer simulation ambiguous. In order to show this, I shall draw a distinction between two possible ways of understanding the notion of simulation, depending on how one views the relation in which a computing system that performs (...) a cognitive task stands to the program that the system runs while performing that task. Next, I shall suggest that the kind of explanation that results from simulation is radically different in each case. In order to illustrate the difference, I will point out some prima facie methodological difficulties that need to be addressed in order to ensure that simulation plays a legitimate explanatory role in cognitive science, and I shall emphasize how those difficulties are very different depending on the notion of explanation involved. (shrink)

Mathematical models are often expected to provide not only predictions about the phenomenon that they represent, but also explanations. These explanations are answers to why-questions and particularly answers to why the predicted phenomenon should occur. For instance, models can be used to calculate when the next total solar eclipse will happen, and then to explain why it will take place on July 2, 2019. In this regard we can obtain explanations from a model if we can solve the model equations (...) which govern the phenomenon under study. But some equations have no explicit solution or are too complicated to solve. In these cases it is difficult for a... (shrink)

Reasons are given to justify the claim that computer simulations and computational science constitute a distinctively new set of scientific methods and that these methods introduce new issues in the philosophy of science. These issues are both epistemological and methodological in kind.

Computational methods such as computer simulations, Monte Carlo methods, and agent-based modeling have become the dominant techniques in many areas of science. Extending Ourselves contains the first systematic philosophical account of these new methods, and how they require a different approach to scientific method. Paul Humphreys draws a parallel between the ways in which such computational methods have enhanced our abilities to mathematically model the world, and the more familiar ways in which scientific instruments have expanded our access to the (...) empirical world. This expansion forms the basis for a new kind of empiricism, better suited to the needs of science than the older anthropocentric forms of empiricism. Human abilities are no longer the ultimate standard of epistemological correctness.Humphreys also includes arguments for the primacy of properties rather than objects, the need to consider technological constraints when appraising scientific methods, and a detailed account of how the path from computational template to scientific application is constructed. This last feature allows us to hold a form of selective realism in which anti-realist arguments based on formal reconstructions of theories can be avoided. One important consequence of the rise of computational methods is that the traditional organization of the sciences is being replaced by an organization founded on computational templates.Extending Ourselves will be of interest to philosophers of science, epistemologists, and to anyone interested in the role played by computers in modern science. (shrink)

I argue that understanding why p involves a kind of intellectual know how and differsfrom both knowledge that p and knowledge why p (as they are standardly understood).I argue that understanding, in this sense, is valuable.

It is often claimed that artificial society simulations contribute to the explanation of social phenomena. At the hand of a particular example, this paper argues that artificial societies often cannot provide full explanations, because their models are not or cannot be validated. Despite that, many feel that such simulations somehow contribute to our understanding. This paper tries to clarify this intuition by investigating whether artificial societies provide potential explanations. It is shown that these potential explanations, if they contribute to our (...) understanding, considerably differ from potential causal explanations. Instead of possible causal histories, simulations offer possible functional analyses of the explanandum . The paper discusses how these two kinds explanatory strategies differ, and how potential functional explanations can be appraised. (shrink)

Among philosophers of science there seems to be a general consensus that understanding represents a species of knowledge, but virtually every major epistemologist who has thought seriously about understanding has come to deny this claim. Against this prevailing tide in epistemology, I argue that understanding is, in fact, a species of knowledge: just like knowledge, for example, understanding is not transparent and can be Gettiered. I then consider how the psychological act of "grasping" that seems to be characteristic of understanding (...) differs from the sort of psychological act that often characterizes knowledge. Zagzebski's account Kvanvig's account Two problems Comanche cases Unreliable sources of information The upper-right quadrant So is understanding a species of knowledge? A false choice. (shrink)

Whereas computer simulations involve no direct physical interaction between the machine they are run on and the physical systems they are used to investigate, they are often used as experiments and yield data about these systems. It is commonly argued that they do so because they are implemented on physical machines. We claim that physicality is not necessary for their representational and predictive capacities and that the explanation of why computer simulations generate desired information about their target system is only (...) to be found in the detailed analysis of their semantic levels. We provide such an analysis and we determine the actual consequences of physical implementation for simulations. (shrink)

Experiments (E), computer simulations (CS) and thought experiments (TE) are usually seen as playing different roles in science and as having different epistemologies. Accordingly, they are usually analyzed separately. We argue in this paper that these activities can contribute to answering the same questions by playing the same epistemic role when they are used to unfold the content of a well-described scenario. We emphasize that in such cases, these three activities can be described by means of the same conceptual framework—even (...) if each of them, because they involve different types of processes, fall under these concepts in different ways. We further illustrate our claims by presenting a threefold case study describing how a TE, a CS and an E were indeed used in the same role at different periods to answer the same questions about the possibility of a physical Maxwellian demon. We also point at fluid dynamics as another field where these activities seem to be playing the same unfolding role. We analyze the importance of unfolding as a general task of science and highlight how our description in terms of epistemic functions articulates in a noncommittal way with the epistemology of these three activities and accounts for their similarities and the existence of hybrid forms of activities. We finally emphasize that picturing these activities as functionally substitutable does not imply that they are epistemologically substitutable. (shrink)

Several philosophical issues in connection with computer simulations rely on the assumption that results of simulations are trustworthy. Examples of these include the debate on the experimental role of computer simulations :483–496, 2009; Morrison in Philos Stud 143:33–57, 2009), the nature of computer data Computer simulations and the changing face of scientific experimentation, Cambridge Scholars Publishing, Barcelona, 2013; Humphreys, in: Durán, Arnold Computer simulations and the changing face of scientific experimentation, Cambridge Scholars Publishing, Barcelona, 2013), and the explanatory power of (...) computer simulations :277–292, 2008; Durán in Int Stud Philos Sci 31:27–45, 2017). The aim of this article is to show that these authors are right in assuming that results of computer simulations are to be trusted when computer simulations are reliable processes. After a short reconstruction of the problem of epistemic opacity, the article elaborates extensively on computational reliabilism, a specified form of process reliabilism with computer simulations located at the center. The article ends with a discussion of four sources for computational reliabilism, namely, verification and validation, robustness analysis for computer simulations, a history of successful implementations, and the role of expert knowledge in simulations. (shrink)

NOW IN PAPERBACK"€"Starting from a collection of simple computer experiments"€"illustrated in the book by striking computer graphics"€"Stephen Wolfram shows how their unexpected results force a whole new way of looking at the operation of our universe.

Verification and validation of numerical models of natural systems is impossible. This is because natural systems are never closed and because model results are always nonunique. Models can be confirmed by the demonstration of agreement between observation and prediction, but confirmation is inherently partial. Complete confirmation is logically precluded by the fallacy of affirming the consequent and by incomplete access to natural phenomena. Models can only be evaluated in relative terms, and their predictive value is always open to question. The (...) primary value of models is heuristic. (shrink)

In this paper, we pursue three general aims: (I) We will define a notion of fundamental opacity and ask whether it can be found in High Energy Physics (HEP), given the involvement of machine learning (ML) and computer simulations (CS) therein. (II) We identify two kinds of non-fundamental, contingent opacity associated with CS and ML in HEP respectively, and ask whether, and if so how, they may be overcome. (III) We address the question of whether any kind of opacity, contingent (...) or fundamental, is unique to ML or CS, or whether they stand in continuity to kinds of opacity associated with other scientific research. (shrink)

Whether simulation models provide the right kind of understanding comparable to that of analytic models has been and remains a contentious issue. The assessment of understanding provided by simulations is often hampered by a conflation between the sense of understanding and understanding proper. This paper presents a deflationist conception of understanding and argues for the need to replace appeals to the sense of understanding with explicit criteria of explanatory relevance and for rethinking the proper way of conceptualizing the role of (...) a single human mind in the collective scientific understanding. (shrink)