Deep neural networks have become increasingly successful in applications from biology to cosmology to social science. Trained DNNs, moreover, correspond to models that ideally allow the prediction of new phenomena. Building in part on the literature on ‘eXplainable AI’, I here argue that these models are instrumental in a sense that makes them non-explanatory, and that their automated generation is opaque in a unique way. This combination implies the possibility of an unprecedented gap between discovery and explanation: When unsupervised models (...) are successfully used in exploratory contexts, scientists face a whole new challenge in forming the concepts required for understanding underlying mechanisms. (shrink)

Scientists regularly make possibility claims. While philosophers of science are well aware of the distinction between epistemic and objective notions of possibility, we believe that they often fail to apply this distinction in their analyses of scientific practices that employ modal concepts. We argue that heeding this distinction will help further progress in current debates in the philosophy of science, as it shows that the debaters talk about different things, rather than disagree on the same issue. We first discuss how (...) the two notions differ with respect to their epistemology and show that these differences are sometimes ignored in the philosophy of science. We then revisit four current philosophy of science debates about modelling, that are framed in modal terms, to showcase how the distinction significantly clarifies these debates and thereby and helps advance them. (shrink)

Computer simulations are involved in numerous branches of modern science, and science would not be the same without them. Yet the question of how they can explain real-world processes remains an issue of considerable debate. In this context, a range of authors have highlighted the inferences back to the world that computer simulations allow us to draw. I will first characterize the precise relation between computer and target of a simulation that allows us to draw such inferences. I then argue (...) that in a range of scientifically interesting cases they are particular abductions and defend this claim by appeal to two case studies. (shrink)

The aim of this article is to use a model from the origin of life studies to provide some depth and detail to our understanding of exploratory models by suggesting that some of these models should be understood as indeterminate. Models that are indeterminate are a type of exploratory model and therefore have extensive potential and can prompt new lines of research. They are distinctive in that, given the current state of scientific understanding, we cannot specify how and where the (...) model will be useful in understanding the natural world: in this case, the origin of life on Earth. The purpose of introducing indeterminacy is to emphasize the epistemic uncertainty associated with modeling, a feature of this practice that has been under emphasized in the literature in favor of attempts to understand the more specific epistemic successes afforded by models. (shrink)

Philosophers of science have recently taken care to highlight different modeling practices where scientific models primarily contribute modal information, in the form of for example possibility claims, how-possibly explanations, or counterfactual conditionals. While examples abound, comparatively little attention is being paid to the question of under what conditions, and in virtue of what, models can perform this epistemic function. In this paper, we firstly delineate modal modeling from other modeling practices, and secondly reviewattempts to spell out and explain the epistemic (...) success conditions of modal modeling. The aim is to more clearly expose the respective justificatory strategies of these accounts, and secondly, to identify lacunae where further work is needed. (shrink)

The ontic conception of explanation, according to which explanations are "full-bodied things in the world," is fundamentally misguided. I argue instead for what I call the eikonic conception, according to which explanations are the product of an epistemic activity involving representations of the phenomena to be explained. What is explained in the first instance is a particular conceptualization of the explanandum phenomenon, contextualized within a given research program or explanatory project. I conclude that this eikonic conception has a number of (...) benefits, including making better sense of scientific practice and allowing for the full range of normative constraints on explanation. (shrink)

Philosophers of science have developed several accounts of how consideration of scientific models can prompt learning about real-world targets. In recent years, various authors advocated the thesis that consideration of so-called minimal models can prompt learning about such targets. In this paper, I draw on the philosophical literature on scientific modelling and on widely cited illustrations from economics and biology to argue that this thesis fails to withstand scrutiny. More specifically, I criticize leading proponents of such thesis for failing to (...) explicate in virtue of what properties or features minimal models supposedly prompt learning and for substantially overstating the epistemic import of minimal models. I then examine and refute several arguments one may put forward to demonstrate that consideration of minimal models can prompt learning about real-world targets. In doing so, I illustrate the implications of my critique for the wider debate on the epistemology of scientific modelling. (shrink)

A primary goal of quantum computer science is to find an explanation for the fact that quantum computers are more powerful than classical computers. In this paper I argue that to answer this question is to compare algorithmic processes of various kinds and to describe the possibility spaces associated with these processes. By doing this, we explain how it is possible for one process to outperform its rival. Further, in this and similar examples little is gained in subsequently asking a (...) how-actually question. Once one has explained how-possibly, there is little left to do. (shrink)

Agent-based modelling has become a well-established method in social epistemology and philosophy of science but the question of what kind of explanations these models provide remains largely open. This paper is dedicated to this issue. It starts by distinguishing between real-world phenomena, real-world possibilities, and logical possibilities as different kinds of targets which agent-based models can represent. I argue that models representing the former two kinds provide how-actually explanations or causal how-possibly explanations. In contrast, models that represent logical possibilities provide (...) epistemically opaque how-possibly explanations. While highly idealised ABMs in the form in which they are initially proposed typically fall into the last category, the epistemic opaqueness of explanations they provide can be reduced by validation procedures. To this purpose, an examination of results of simulations in terms of classes of models can be particularly helpful. I illustrate this point by discussing a class of ABMs of scientific interaction and the claim that a high degree of interaction can impede scientific inquiry. (shrink)

‘Possibilist Explanation’ is a promising account of scientific explanation which avoids the familiar problems of “how-possibly explanations”. It explains an event by showing how-actually it was epistemically possible, instead of why it was epistemically necessary. Its explanandum is the epistemic possibility of an actual event previously considered epistemically impossible. To define PE, two new concepts are introduced: ‘permissive condition’ and ‘possibilist law’. A permissive condition for an event is something that does not entail the event itself, but a necessary condition (...) for it. A ‘Possibilist Law’ is a kind of scientific law that predicts that in the absence of a necessary condition N, the event E is not possible. Both PE and PL are legitimate and neglected parts of scientific knowledge and are especially suitable for human sciences. (shrink)

In this essay, I provide an overview of the debate on infinite and essential idealizations in physics. I will first present two ostensible examples: phase transitions and the Aharonov– Bohm effect. Then, I will describe the literature on the topic as a debate between two positions: Essentialists claim that idealizations are essential or indispensable for scientific accounts of certain physical phenomena, while dispensabilists maintain that idealizations are dispensable from mature scientific theory. I will also identify some attempts at finding a (...) middle ground between the essentialists and dispensabilists camps. Finally, I will raise questions for future research on essential and infinite idealizations via the notion of exploration. (shrink)

How can we use models to understand real phenomena if models misrepresent the very phenomena we seek to understand? Some accounts suggest that models may afford understanding by providing causal knowledge about phenomena via how-possibly explanations. However, general equilibrium models, for example, pose a challenge to this solution since their contribution appears to be purely mathematical results. Despite this, practitioners widely acknowledge that it improves our understanding of the world. I argue that the Arrow–Debreu model provides a mathematical how-possibly explanation (...) which establishes claims of mathematical dependence. The account developed reveals how mathematical knowledge can inform claims about the world, allow ‘what-if-things-had-been-different’ inferences, and thus improve our understanding. (shrink)

A recent series of experiments have demonstrated that a classical fluid mechanical system, constituted by an oil droplet bouncing on a vibrating fluid surface, can be induced to display a number of behaviours previously considered to be distinctly quantum. To explain this correspondence it has been suggested that the fluid mechanical system provides a single-particle classical model of de Broglie’s idiosyncratic ‘double solution’ pilot wave theory of quantum mechanics. In this paper we assess the epistemic function of the bouncing oil (...) droplet experiments in relation to quantum mechanics. We find that the bouncing oil droplets are best conceived as an analogue illustration of quantum phenomena, rather than an analogue simulation, and, furthermore, that their epistemic value should be understood in terms of how-possibly explanation, rather than confirmation. Analogue illustration, unlike analogue simulation, is not a form of ‘material surrogacy’, in which source empirical phenomena in a system of one kind can be understood as ‘standing in for’ target phenomena in a system of another kind. Rather, analogue illustration leverages a correspondence between certain empirical phenomena displayed by a source system and aspects of the ontology of a target system. On the one hand, this limits the potential inferential power of analogue illustrations, but, on the other, it widens their potential inferential scope. In particular, through analogue illustration we can learn, in the sense of gaining how-possibly understanding, about the putative ontology of a target system via an experiment. As such, the potential scientific value of these extraordinary experiments is undoubtedly a significant one. (shrink)

One puzzle concerning highly idealized models is whether they explain. Some suggest they provide so-called ‘how-possibly explanations’. However, this raises an important question about the nature of how-possibly explanations, namely what distinguishes them from ‘normal’, or how-actually, explanations? I provide an account of how-possibly explanations that clarifies their nature in the context of solving the puzzle of model-based explanation. I argue that the modal notions of actuality and possibility provide the relevant dividing lines between how-possibly and how-actually explanations. Whereas how-possibly (...) explanations establish claims of possible explanations, how-actually explanations establish claims of actual ones. Models, in turn, simply provide evidence for these claims. (shrink)

In this paper we examine the epistemic value of highly idealized agent-based models of social aspects of scientific inquiry. On the one hand, we argue that taking the results of such simulations as informative of actual scientific inquiry is unwarranted, at least for the class of models proposed in recent literature. Moreover, we argue that a weaker approach, which takes these models as providing only “how-possibly” explanations, does not help to improve their epistemic value. On the other hand, we suggest (...) that if ABMs of science underwent two types of robustness analysis, they could indeed have a clear epistemic function, namely by providing evidence for philosophical and historical hypotheses. In this sense, ABMs can obtain evidential and explanatory properties and thus be a useful tool for integrated history and philosophy of science. We illustrate our point with an example of a model—building on the work by Kevin Zollman—which we apply to a concrete historical case study. (shrink)

In Simulation and Similarity, Michael Weisberg offers a similarity-based account of the model–world relation, which is the relation in virtue of which successful models are successful. Weisberg’s main idea is that models are similar to targets in virtue of sharing features. An important concern about Weisberg’s account is that it remains silent on what it means for models and targets to share features, and consequently on how feature-sharing contributes to models’ epistemic success. I consider three potential ways of concretizing the (...) concept of shared features: as identical, quantitatively sufficiently close, and sufficiently similar features. I argue that each of these concretizations faces significant challenges, leaving unclear how Weisberg’s account substantially contributes to elucidating the relation in virtue of which successful models are successful. Against this background, I outline a pluralistic revision and argue that this revision may not only help Weisberg's account evade several of the problems that I raise, but also offers a novel perspective on the model–world relation more generally. 1Introduction 2Weisberg’s Feature-Sharing Account 3What Is a Shared Feature? 3.1Identity 3.2Sufficient closeness 3.3Sufficient similarity 4Turning Weisberg’s Account ‘Upside Down’ 5Conclusion. (shrink)

Science is replete with falsehoods that epistemically facilitate understanding by virtue of being the very falsehoods they are. In view of this puzzling fact, some have relaxed the truth requirement on understanding. I offer a factive view of understanding that fully accommodates the puzzling fact in four steps: (i) I argue that the question how these falsehoods are related to the phenomenon to be understood and the question how they figure into the content of understanding it are independent. (ii) I (...) argue that the falsehoods do not figure into the understanding’s content by being elements of its periphery or core. (iii) Drawing lessons from case studies, I argue that the falsehoods merely enable understanding. When working with such falsehoods, only the truths we extract from them are elements of the content of our understanding. (iv) I argue that the extraction view is compatible with the thesis that falsehoods can have an epistemic value by virtue of being the very falsehoods they are. (shrink)

One can find in the literature two sets of views concerning the relationship between understanding and explanation: that one understands only if 1) one has knowledge of causes and 2) that knowledge is provided by an explanation. Taken together, these tenets characterize what I call the narrow knowledge account of understanding. While the first tenet has recently come under severe attack, the second has been more resistant to change. I argue that we have good reasons to reject it on the (...) basis of theoretical models that provide how-possibly explanations. These models, while they do not explain in the strict sense, afford understanding. In response, I propose an alternative epistemology of understanding, broad KAU, that takes cases of theoretical modelling into account. (shrink)

The question of whether the idealized models of theoretical economics are explanatory has been the subject of intense philosophical debate. It is sometimes presupposed that either a model provides the actual explanation or it does not provide an explanation at all. Yet, two sets of issues are relevant to the evaluation of model-based explanation: what conditions should a model satisfy in order to count as explanatory and does the model satisfy those conditions. My aim in this paper is to unpack (...) this distinction and show that separating the first set of issues from the second is crucial to an accurate diagnosis of the distinctive challenges that economic models pose. Along the way I sketch a view of model-based explanation in economics that focuses on the role that non-empirical and empirical strategies play in increasing confidence in the adequacy of a given model-based explanation. (shrink)

In the context of astrophysical modeling at the solar system scale, we investigate the modalities implied by taking into account different levels of detail at which phenomena can be considered. In particular, by framing the analysis in terms of the how-possibly/how-actually distinction, we address the debated question as to whether the degree of plausibility is tightly linked to the degree of detail. On the grounds of concrete examples, we argue that, also in the astrophysical context examined, this is not necessarily (...) the case. (shrink)