It is a common view among philosophers of science that theoretical virtues (also known as epistemic or cognitive values), such as simplicity and consistency, play an important role in scientific practice. In this article, I set out to study the role that theoretical virtues play in scientific practice empirically. I apply the methods of data science, such as text mining and corpus analysis, to study large corpora of scientific texts in order to uncover patterns of usage. These patterns of usage, (...) in turn, might shed some light on the role that theoretical virtues play in scientific practice. Overall, the results of this empirical study suggest that scientists invoke theoretical virtues explicitly, albeit rather infrequently, when they talk about models (less than 30%), theories (less than 20%), and hypotheses (less than 15%) in their published works. To the extent that they are mentioned in scientific publications, the results of this study suggest that accuracy, consistency, and simplicity are the theoretical virtues that scientists invoke more frequently than the other theoretical virtues tested in this study. Interestingly, however, depending on whether they talk about hypotheses, theories, or models, scientists may invoke one of those theoretical virtues more than the others. (shrink)

The decision whether to have a realist or an anti-realist attitude towards scientific hypotheses is interpreted in this paper as a choice that scientists themselves have to face in their work as scientists, rather than as a ‘philosophical’ problem. Scientists’ choices between realism and instrumentalism are interpreted in this paper with the help of two different conceptual tools: a deflationary semantics grounded in the inferentialist approach to linguistic practices developed by some authors, and an epistemic utility function that tries to (...) represent the cognitive preferences of scientists. The first tool is applied to two different questions traditionally related to the problem of scientific realism: the non-miracle argument, and the continuity of reference. The second one is applied to the problem of unconceived alternatives, and to the distinction between realist and instrumentalist attitudes towards scientific hypotheses. (shrink)

We have previously argued that historical cases must be rendered canonical before they can plausibly serve as evidence for philosophical claims, where canonicity is established through a process of negotiation among historians and philosophers of science (Bolinska and Martin, 2020). Here, we extend this proposal by exploring how that negotiation might take place in practice. The working stock of historical examples that philosophers tend to employ has long been established informally, and, as a result, somewhat haphazardly. The composition of the (...) historical canon of philosophy of science is therefore path dependent, and cases often become stock examples for reasons tangential to their appropriateness for the purposes at hand. We show how the lack of rigor around the canonization of case studies has muddied the waters in selected philosophical debates. This, in turn, lays the groundwork for proposing ways in which they can be improved. (shrink)

Neuroscientists have in recent years turned to building models that aim to generate predictions rather than explanations. This “predictive turn” has swept across domains including law, marketing, and neuropsychiatry. Yet the norms of prediction remain undertheorized relative to those of explanation. I examine two styles of predictive modeling and show how they exemplify the normative dynamics at work in prediction. I propose an account of how predictive models, conceived of as technological devices for aiding decision-making, can come to be adequate (...) for purposes that are defined by both their guiding research questions and their larger social context of application. (shrink)

Stanford’s argument against scientific realism focuses on theories, just as many earlier arguments from inconceivability have. However, there are possible arguments against scientific realism involving unconceived (or inconceivable) entities of different types: observations, models, predictions, explanations, methods, instruments, experiments, and values. This paper charts such arguments. In combination, they present the strongest challenge yet to scientific realism.

In their attempt to defend philosophy from accusations of uselessness made by prominent scientists, such as Stephen Hawking, some philosophers respond with the charge of ‘scientism.’ This charge makes endorsing a scientistic stance, a mistake by definition. For this reason, it begs the question against these critics of philosophy, or anyone who is inclined to endorse a scientistic stance, and turns the scientism debate into a verbal dispute. In this paper, I propose a different definition of scientism, and thus a (...) new way of looking at the scientism debate. Those philosophers who seek to defend philosophy against accusations of uselessness would do philosophy a much better service, I submit, if they were to engage with the definition of scientism put forth in this paper, rather than simply make it analytic that scientism is a mistake. (shrink)

I present a recent historical case from cosmology—the story of inflationary cosmology—and on its basis argue that solving explanatory problems is a reliable method for making progress in science. In particular, I claim that the success of inflationary theory at solving its predecessor’s explanatory problems justified the theory epistemically, even in advance of the development of novel predictions from the theory and the later confirmation of those predictions.

The twentieth century enjoys a firm grip on our profession. Well over half the research articles published in this journal since 2000 devote significant attention to the period between the 1890s and the 1990s. Similar trends prevail in other leading publications. But this outpouring of scholarship alone does not create a collective sense of how historians of science should confront the twentieth century as an epoch. The synthetic reflection that established the scientific revolution as a historiographical category and lent the (...) nineteenth century a sense of cohesion remains to be undertaken for the twentieth. A panoramic outlook on more recent historical eras is a pressing necessity. Publishing trends suggest that we will confront future methodological questions and navigate our evolving professional identity largely on twentieth-century turf. What, then, defines twentieth-century science? In some ways, this question is naive, even trivial: it surprises no one that the currents of history disdain our thin calendrical embankments. But centuries, decades, and other conventional durations often prove useful for collating practices that shared a family resemblance and naming periods about which we might claim that continuity prevailed over discontinuity. Can the twentieth century offer such utility? The books under review can each be read as an attempt to use the twentieth century in that way and, in so doing, to claim it for a particular vision of how the history of science should be done. (shrink)

In the last few years, biologists and computer scientists have claimed that the introduction of data science techniques in molecular biology has changed the characteristics and the aims of typical outputs (i.e. models) of such a discipline. In this paper we will critically examine this claim. First, we identify the received view on models and their aims in molecular biology. Models in molecular biology are mechanistic and explanatory. Next, we identify the scope and aims of data science (machine learning in (...) particular). These lie mainly in the creation of predictive models which performances increase as data set increases. Next, we will identify a tradeoff between predictive and explanatory performances by comparing the features of mechanistic and predictive models. Finally, we show how this a priori analysis of machine learning and mechanistic research applies to actual biological practice. This will be done by analyzing the publications of a consortium—The Cancer Genome Atlas—which stands at the forefront in integrating data science and molecular biology. The result will be that biologists have to deal with the tradeoff between explaining and predicting that we have identified, and hence the explanatory force of the ‘new’ biology is substantially diminished if compared to the ‘old’ biology. However, this aspect also emphasizes the existence of other research goals which make predictive force independent from explanation. (shrink)

In the last few years, biologists and computer scientists have claimed that the introduction of data science techniques in molecular biology has changed the characteristics and the aims of typical outputs (i.e. models) of such a discipline. In this paper we will critically examine this claim. First, we identify the received view on models and their aims in molecular biology. Models in molecular biology are mechanistic and explanatory. Next, we identify the scope and aims of data science (machine learning in (...) particular). These lie mainly in the creation of predictive models which performances increase as data set increases. Next, we will identify a tradeoff between predictive and explanatory performances by comparing the features of mechanistic and predictive models. Finally, we show how this a priori analysis of machine learning and mechanistic research applies to actual biological practice. This will be done by analyzing the publications of a consortium—The Cancer Genome Atlas—which stands at the forefront in integrating data science and molecular biology. The result will be that biologists have to deal with the tradeoff between explaining and predicting that we have identified, and hence the explanatory force of the ‘new’ biology is substantially diminished if compared to the ‘old’ biology. However, this aspect also emphasizes the existence of other research goals which make predictive force independent from explanation. (shrink)

The predictive inadequacy of the social sciences is well documented, and philosophers have sought to diagnose it. This paper examines Brian Epstein’s recent diagnosis. He argues that the social sciences treat the social world as entirely composed of individual people. Instead, social scientists should recognize that material, non-individualistic entities determine the social world, as well. First, I argue that Epstein’s argument both begs the question against his opponents and is not sufficiently charitable. Second, I present doubts that his proposal will (...) improve predictive success for the social sciences, which I support with Edith Penrose’s resource-based theory of the firm. (shrink)

There are at least twelve major virtues of good theories: evidential accuracy, causal adequacy, explanatory depth, internal consistency, internal coherence, universal coherence, beauty, simplicity, unification, durability, fruitfulness, and applicability. These virtues are best classified into four classes: evidential, coherential, aesthetic, and diachronic. Each virtue class contains at least three virtues that sequentially follow a repeating pattern of progressive disclosure and expansion. Systematizing the theoretical virtues in this manner clarifies each virtue and suggests how they might have a coordinated and cumulative (...) role in theory formation and evaluation across the disciplines—with allowance for discipline specific modification. An informal and flexible logic of theory choice is in the making here. Evidential accuracy, according to my systematization, is not a largely isolated trait of good theories, as some have made it out to be. Rather, it bears multifaceted relationships, constituting significant epistemic entanglements, with other theoretical virtues. (shrink)

We provide a novel articulation of the epistemic peril of p-hacking using three resources from philosophy: predictivism, Bayesian confirmation theory, and model selection theory. We defend a nuanced position on p-hacking: p-hacking is sometimes, but not always, epistemically pernicious. Our argument requires a novel understanding of Bayesianism, since a standard criticism of Bayesian confirmation theory is that it cannot represent the influence of biased methods. We then turn to pre-analysis plans, a methodological device used to mitigate p-hacking. Some say that (...) pre-analysis plans are epistemically meritorious while others deny this, and in practice pre-analysis plans are often violated. We resolve this debate with a modest defence of pre-analysis plans. Further, we argue that pre-analysis plans can be epistemically relevant even if the plan is not strictly followed—and suggest that allowing for flexible pre-analysis plans may be the best available policy option. (shrink)

Scientific hypotheses may either predict particular unknown facts or accommodate previously-known data. Although affirmed predictions are intuitively more rewarding than accommodations of established facts, opinions divide whether predictive hypotheses are also epistemically superior to accommodation hypotheses. This paper examines the contribution of predictive hypotheses to discoveries of several bio-molecular systems. Having all the necessary elements of the system known beforehand, an abstract predictive hypothesis of semiconservative mode of DNA replication was successfully affirmed. However, in defining the genetic code whose biochemical (...) basis was unclear, hypotheses were only partially effective and supplementary experimentation was required for its conclusive definition. Markedly, hypotheses were entirely inept in predicting workings of complex systems that included unknown elements. Thus, hypotheses did not predict the existence and function of mRNA, the multiple unidentified components of the protein biosynthesis machinery, or the manifold unknown constituents of the ubiquitin–proteasome system of protein breakdown. Consequently, because of their inability to envision unknown entities, predictive hypotheses did not contribute to the elucidation of cation theories remained the sole instrument to explain complex bio-molecular systems, the philosophical question of alleged advantage of predictive over accommodative hypotheses became inconsequential. (shrink)

Many climate scientists have made claims that may suggest that evidence used in tuning or calibrating a climate model cannot be used to evaluate the model. By contrast, the philosophers Katie Steele and Charlotte Werndl have argued that, at least within the context of Bayesian confirmation theory, tuning is simply an instance of hypothesis testing. In this paper I argue for a weak predictivism and in support of a nuanced reading of climate scientists’ concerns about tuning: there are cases, model-tuning (...) among them, in which predictive successes are more highly confirmatory of a model than accommodation of evidence. (shrink)

There has been a lively debate in the philosophy of science over _predictivism_: the thesis that successfully predicting a given body of data provides stronger evidence for a theory than merely accommodating the same body of data. I argue for a very strong version of the thesis using statistical results on the so-called “model selection” problem. This is the problem of finding the optimal model (family of hypotheses) given a body of data. The key idea that I will borrow from (...) the statistical literature is that the level of support a hypothesis, _H,_ receives from a body of data, _D,_ is inversely related to the number of adjustable parameters of the model from which _H_ was constructed. I will argue that when _D_ is not essential to the design of _H_ (i.e., when it is predicted), the model to which _H_ belongs has fewer adjustable parameters than when _D_ is essential to the design of _H_ (when it is accommodated)_._ This, I argue, provides us with an argument for a very strong version of predictivism. (shrink)

Thought experiments (TEs) are important tools in science, used to both undermine and support theories, and communicate and explain complex phenomena. Their interest within philosophy of science has been dominated by a narrow question: How do TEs increase knowledge? My aim is to push beyond this to consider their broader value in scientific practice. I do this through an investigation into the scientific imagination. Part one explores questions regarding TEs as “experiments in the imagination” via a debate concerning the epistemic (...) status of computer simulations in science. I outline how, against Hacking, TEs also have “a life of their own” and I argue against accounts that privilege experiments over simulations (and by extension TEs) in light of their capacity to surprise in a productive way. Part two develops a pluralist account of the nature of the imagination in science. At its core, my view is that when we attend to a number of examples of TEs and consider the context in which they are used, we see that TEs engage a variety of our imaginative capacities. Existing monistic views fail to recognise the richness of the imagination and its potential in science. Part three looks to the “beauty” of TEs which is currently overlooked in the aesthetics of science literature. I put forward a new account that demonstrates the epistemic value of aesthetic features in science by showing how an appropriate fit between form and content enhances the usability of a TE, and its effectiveness as a prompt for our imagination. This also enables a more nuanced take on the proposed similarities between TEs and literary fictions. In the concluding chapter, I outline ways in which the core features of my account can be extended beyond TEs to illuminate the significance of the imagination and aesthetic values in other areas of science. (shrink)

Heather Douglas has helped to set the standard for twenty-first century discussions in philosophy of science on the topics of values in science and science in democracy. Douglas’s work has been part of a movement to bring the question of values in science back to center of the field and to focus especially on policy-relevant science. This first chapter, on the pervasive entanglement of science and values, includes an improved and definitive statement of the argument from inductive risk, which she (...) is single-handedly responsible for rehabilitating and returning to the center of the debate. This statement makes clear the fundamental and absolutely pervasive nature of inductive risk and its import for our understanding of the role of values in science. The chapter also provides a survey of the current field of alternative approaches to ideals for the epistemic role of values in science that is comprehensive and generous, yet critical of each. -/- In these brief comments, I focus on providing an alternative perspective on some conceptual and rhetorical issues in Douglas’s account, specifically dealing with the nature of values and the relation of the descriptive and the normative. This will lead me to somewhat different evaluations of two of the five new ideals for values in science that Douglas canvasses in the chapter. (shrink)

Many philosophers have argued that a hypothesis is better confirmed by some data if the hypothesis was not specifically designed to fit the data. ‘Prediction’, they argue, is superior to ‘accommodation’. Others deny that there is any epistemic advantage to prediction, and conclude that prediction and accommodation are epistemically on a par. This paper argues that there is a respect in which accommodation is superior to prediction. Specifically, the information that the data was accommodated rather than predicted suggests that the (...) data is less likely to have been manipulated or fabricated, which in turn increases the likelihood that the hypothesis is correct in light of the data. In some cases, this epistemic advantage of accommodation may even outweigh whatever epistemic advantage there might be to prediction, making accommodation epistemically superior to prediction all things considered. (shrink)

In the last two decades, philosophy of neuroscience has predominantly focused on explanation. Indeed, it has been argued that mechanistic models are the standards of explanatory success in neuroscience over, among other things, topological models. However, explanatory power is only one virtue of a scientific model. Another is its predictive power. Unfortunately, the notion of prediction has received comparatively little attention in the philosophy of neuroscience, in part because predictions seem disconnected from interventions. In contrast, we argue that topological predictions (...) can and do guide interventions in science, both inside and outside of neuroscience. Topological models allow researchers to predict many phenomena, including diseases, treatment outcomes, aging, and cognition, among others. Moreover, we argue that these predictions also offer strategies for useful interventions. Topology-based predictions play this role regardless of whether they do or can receive a mechanistic interpretation. We conclude by making a case for philosophers to focus on prediction in neuroscience in addition to explanation alone. (shrink)

The practice of HARKing---hypothesizing after results are known---is commonly maligned as undermining the reliability of scientific findings. There are several accounts in the literature as to why HARKing undermines the reliability of findings. We argue that none of these is right and that the correct account is a Bayesian one. HARKing can indeed decrease the reliability of scientific findings, but it can also increase it; which effect HARKing produces depends on the difference of the prior odds of hypotheses characteristically selected (...) ex ante and ex post to observing data. Further, we show how misdiagnosis of HARKing can lead to misprescription in the context of the replication crisis. (shrink)