Science promises benefits, some true and some illusory. Consider a scientific agnostic who thinks that to reap the true benefits of a scientific theory he does not have to believe in its theoretical posits. Instead, it is enough if he believes that the theory successfully predicts the behavior of the observables, as ultimately only such predictions matter. -/- Say, however, that given the results of her thorough research, a psychologist proposes a theory describing a psychological mechanism underlying a certain class (...) of normative judgments. Moreover, the mechanism seems unfit for the task—once you see the details of the mechanism, you will realize that this is not the way they should be produced. Therefore, if the psychologist is right, it seems that these normative judgments should not inform one's normative theorizing or one's actions (Greene 2008, 2014, Kelly 2014). And say that the agnostic accepts the psychologist's theory, trusting that it makes correct predictions about, for example, fMRI images and subjects' reaction times, as they are observable. He also thinks that if the psychologist's description of the mechanism is correct, the judgments should not be trusted. Yet, since the mechanism posited by the theory is not observable, the agnostic is agnostic about it. He thus cannot be convinced that these judgments are produced in a flawed way and, consequently, has no reason to distrust them. -/- Scientific agnosticism comes, therefore, at the cost of dismissing normative arguments that invoke unobservable posits of psychological models. The ability to make such arguments is a true (rather than illusory) benefit of science, despite the agnostic's promise that his philosophical theory leaves intact benefits that genuinely matter. (shrink)

This article argues that a successful answer to Hume's problem of induction can be developed from a sub-genre of philosophy of science known as formal learning theory. One of the central concepts of formal learning theory is logical reliability: roughly, a method is logically reliable when it is assured of eventually settling on the truth for every sequence of data that is possible given what we know. I show that the principle of induction (PI) is necessary and sufficient for logical (...) reliability in what I call simple enumerative induction. This answer to Hume's problem rests on interpreting PI as a normative claim justified by a non-empirical epistemic means-ends argument. In such an argument, a rule of inference is shown by mathematical or logical proof to promote a specified epistemic end. Since the proof concerning PI and logical reliability is not based on inductive reasoning, this argument avoids the circularity that Hume argued was inherent in any attempt to justify PI. (shrink)

The no-free-lunch theorems promote a skeptical conclusion that all possible machine learning algorithms equally lack justification. But how could this leave room for a learning theory, that shows that some algorithms are better than others? Drawing parallels to the philosophy of induction, we point out that the no-free-lunch results presuppose a conception of learning algorithms as purely data-driven. On this conception, every algorithm must have an inherent inductive bias, that wants justification. We argue that many standard learning algorithms should rather (...) be understood as model-dependent: in each application they also require for input a model, representing a bias. Generic algorithms themselves, they can be given a model-relative justification. (shrink)

Putnam construed the aim of Carnap’s program of inductive logic as the specification of a “universal learning machine,” and presented a diagonal proof against the very possibility of such a thing. Yet the ideas of Solomonoff and Levin lead to a mathematical foundation of precisely those aspects of Carnap’s program that Putnam took issue with, and in particular, resurrect the notion of a universal mechanical rule for induction. In this paper, I take up the question whether the Solomonoff–Levin proposal is (...) successful in this respect. I expose the general strategy to evade Putnam’s argument, leading to a broader discussion of the outer limits of mechanized induction. I argue that this strategy ultimately still succumbs to diagonalization, reinforcing Putnam’s impossibility claim. (shrink)

A careful analysis of Salmon’s Theoretical Realism and van Fraassen’s Constructive Empiricism shows that both share a common origin: the requirement of literal construal of theories inherited by the Standard View. However, despite this common starting point, Salmon and van Fraassen strongly disagree on the existence of unobservable entities. I argue that their different ontological commitment towards the existence of unobservables traces back to their different views on the interpretation of probability via different conceptions of induction. In fact, inferences to (...) statements claiming the existence of unobservable entities are inferences to probabilistic statements, whence the crucial importance of the interpretation of probability. (shrink)

Take the following version of scientific realism: we have good reason to believe that (some of the) current scientific theories tell us something specific about the underlying, i.e. unobservable, structures of the world, for instance that there are electrons with a certain electric charge, or that there are viruses that cause certain diseases. Popper, the rationalist, would not have adhered to the proposed formulation of scientific realism in terms of the rationality of existential beliefs concerning unobservables. Popper did not believe (...) in belief. According to Van Fraassen, the empiricist, one may yet have a rational existential belief concerning unobservables, given a liberal notion of rationality of belief. In this paper I will investigate to what extent a reassessment of both Popper’s rejection of the rationality of belief and Van Fraassen’s reformulation of the rationality of belief, points towards a new and pragmatist dissolution of the ‘problem of scientific realism’. (shrink)

This paper connects veritistic teleological epistemology, VTE, with the epistemological dimension of the scientific realism debate. VTE sees our epistemic activities as a tradeoff between believing truths and avoiding error. I argue that van Fraassen’s epistemology is not suited to give a justification for a crucial presupposition of his Bad Lot objection to inference to the best explanation (IBE), the presupposition that believing that p is linked to p being more likely to be true. This makes him vulnerable to a (...) counterargument, tailored after Musgrave’s defense of IBE, which would result in a stalemate between them about presuppositions of rationality. I will, however, show that switching to VTE can justify van Fraassen’s presupposition. This leads to a dismissal of common IBE arguments for realism as presented by Boyd and Musgrave, but I also argue that a more cautious version of realism can be rescued from the Bad Lot objection. Finally, I raise some worries about epistemic risk-attitude consistency for constructive empiricists and develop an alternative anti-realist position.1. (shrink)

A traditional view is that to be an empiricist is to hold a particular epistemological belief: something to the effect that knowledge must derive from experience. In his recent book The Empirical Stance, and in a number of other publications, Bas van Fraassen has disagreed, arguing that if empiricism is to be defensible it must instead be thought of as a stance: an attitude of mind or methodological orientation rather than a factual belief. In this article I will examine his (...) arguments for this claim in detail. I will argue that they do not succeed and that empiricism is, contrary to van Fraassen's claim, better thought of as a truth‐evaluable doctrine than as a stance. (shrink)

Many philosophers agree that Hume was not simply objecting to inductive inferences on the grounds of their logical invalidity and that his description of our inductive behaviour was inadequate, but none the less regard his argument against induction as irrefutable. I argue that this constellation of opinions contains a serious tension. In the light of the tension, I re-examine Hume’s actual sceptical argument and show that the argument as it stands is valid but unsound. I argue that it can only (...) be converted into a sound one if our inductive behaviour can be characterized as a process of rule-governed ampliation. Drawing on some Bayesian ideas, I argue that our inductive behaviour probably cannot be characterized in that way, so our immunity from Hume is secure. Finally, I compare my response to Hume’s argument with some other well known responses. (shrink)

A standard way to challenge convergence-based accounts of inductive success is to claim that they are too weak to constrain inductive inferences in the short run. We respond to such a challenge by answering some questions raised by Juhl (1994). When it comes to predicting limiting relative frequencies in the framework of Reichenbach, we show that speed-optimal convergence—a long-run success condition—induces dynamic coherence in the short run.

In this paper, I argue that the “positive argument” for Constructive Empiricism (CE), according to which CE “makes better sense of science, and of scientific activity, than realism does” (van Fraassen 1980, 73), is an Inference to the Best Explanation (IBE). But constructive empiricists are critical of IBE, and thus they have to be critical of their own “positive argument” for CE. If my argument is sound, then constructive empiricists are in the awkward position of having to reject their own (...) “positive argument” for CE by their own lights. (shrink)

In this paper, I argue that Constructive Empiricism (CE) is ambiguous between two interpretations: CE as a normative epistemology of science and CE as a descriptive philosophy of science. When they present CE, constructive empiricists write as if CE is supposed to be more than a normative epistemology of science and that it is meant to be responsible to actual scientific practices. However, when they respond to objections, constructive empiricists fall back on a strictly normative interpretation of CE. This ambiguity (...) seems to make CE immune to objections in a rather ad hoc fashion. (shrink)

Abstractvan Fraassen (The empirical stance, 2002) contrasts the empirical stance with the materialist stance. The way he describes them makes both of them attractive, and while opposed they have something in common for both stances are scientific approaches to philosophy. The difference between them reflects their differing conceptions of science itself. Empiricists emphasise fallibilism, verifiability and falsifiability, and also to some extent scepticism and tolerance of novel hypotheses. Materialists regard the theoretical picture of the world as matter in motion as (...) a true and explanatory account and insist on not taking ‘spooky’ entities or processes seriously as potential explanations of phenomena that so far lie outside the scope of successful science. The history of science shows us that both stances have been instrumental in the achievement of progress at various times. It is therefore plausible for a naturalist to suggest that science depends for its success on the dialectic between empiricism and materialism. A truly naturalist approach to philosophy ought then to synthesise them. Call the synthesized empiricist and materialist stances ‘the scientistic stance’. This paper elaborates and defends it. (shrink)

The question of the role of theory in the determination of reference of theoretical terms continues to be a controversial one. In the present paper I assess a number of responses to this question (including variations on David Lewis’s appeal to Ramsification), before describing an alternative, epistemically oriented account of the reference-determination of such terms. The paper concludes by discussing some implications of the account for our understanding of both realism and such competitors of realism as constructive empiricism.

It has become apparent that the debate between scientific realists and constructive empiricists has come to a stalemate. Neither view can reasonably claim to be the most rational philosophy of science, exclusively capable of making sense of all scientific activities. On one prominent analysis of the situation, whether we accept a realist or an anti-realist account of science actually seems to depend on which values we antecedently accept, rather than our commitment to “rationality” per se. Accordingly, several philosophers have attempted (...) to argue in favour of scientific realism or constructive empiricism by showing that one set of values is exclusively best, for anyone and everyone, and that the downstream choice of the philosophy of science which best serves those values is therefore best, for anyone and everyone. These efforts, however, seem to have failed. In response, I suggest that philosophers of science should suspend the effort to determine which philosophy of science is best for everyone, and instead begin investigating which philosophy of science is best for specific people, with specific values, in specific contexts. I illustrate how this might be done by briefly sketching a single case study from the history of science, which seems to show that different philosophies of science are better at motivating different forms of scientific practice. (shrink)

In the literature, one finds two accounts of the normative status of rational belief: the ought account and the permissibility account. Both accounts have their advantages and shortcomings, making it difficult to favour one over the other. Imagine that there were two principles of rational belief or rational degrees of belief commonly considered plausible, but which, however, yielded a paradox together with one account, but not with the other. One of the accounts therefore requires us to give up one of (...) the plausible principles; whereas the other allows us to save them both. The fact that it allows us to save both of the plausible principles might well be considered a strong reason in favour of the relevant account. The permissibility-account-based resolution of the lottery paradox suggests that the permissibility account is a candidate for being supported in this way, since the account seems to save two plausible principles of rational belief and rational degrees of belief. I argue that even if the permissibility account were supported in this way the support would be defeated, since one cannot provide an analogous resolution of the preface paradox. The principles remain unsaved by the permissibility account. (shrink)

SummaryTake the following version of scientific realism: we have good reason to believe that (some of the) current scientific theories tell us something specific about the underlying, i.e. unobservable, structures of the world, for instance that there are electrons with a certain electric charge, or that there are viruses that cause certain diseases. Popper, the rationalist, would not have adhered to the proposed formulation of scientific realism in terms of the rationality of existential beliefs concerning unobservables. Popper did not believe (...) in belief. According to Van Fraassen, the empiricist, one may yet have a rational existential belief concerning unobservables, given a liberal notion of rationality of belief. In this paper I will investigate to what extent a reassessment of both Popper’s rejection of the rationality of belief and Van Fraassen’s reformulation of the rationality of belief, points towards a new and pragmatist dissolution of the ‘problem of scientific realism’. (shrink)

According to Bas van Fraassen, scientific realists and anti-realists disagree about whether accepting a scientific theory involves believing that the theory is true. On van Fraassen’s own anti-realist empiricist position, accepting a theory involves believing only that the theory is correct in its claims about observable aspects of the world. However, a number of philosophers have argued that acceptance and belief cannot be distinguished and thus that the debate is either confused or trivially settled in favor of the realist. In (...) addition, another set of philosophers have argued that van Fraassen’s empiricist position appeals to an unmotivated distinction between observable and unobservable aspects of the world. This paper aims to reconstruct a van Fraassen-style empiricism about scientific acceptance that avoids these two objections – reconstructed empiricism. (shrink)

Voluntarism about beliefs is the view that persons can be free to choose their beliefs for non-epistemic (truth-related) reasons (cf. Williams 1973). One problem for belief voluntarism is that it can lead to Moore-paradoxality. The person might believe that -/- a.) there are also good epistemic reasons for her belief, or that b.) there are no epistemic reasons one way or the other, or that c.) there are good epistemic reasons against her belief. -/- If the person is aware of (...) the fact that she chose her belief for non-epistemic reasons, then she would believe one of the following three things (using "evidence" as a term for all kinds of epistemic reasons): -/- a.) I believe that p but not because of the good evidence in favour of it; b.) I believe that p but that has nothing to do with the evidence; c.) I believe that p but the evidence speaks against it. -/- We could add a fourth case in which the person suspends belief on the subject matter despite the evidence: -/- d.) The evidence supports "p" but I have no view on the matter. -/- All the different versions a-d are somehow Moore-paradoxical. They constitute different ways of mistrusting one's own belief (cf. Wittgenstein 1958, p.190). Given that one cannot, for conceptual reasons, mistrust one's beliefs, it follows that a person cannot be in one of the above predicaments described by a-d. Hence, voluntarism about beliefs is false, at least if we are dealing with cases in which the person is aware of the fact that she chose her belief for non-epistemic reasons. It won't help to focus on cases in which the person is not aware of that fact, either because she has forgotten the genesis of her belief or because the process of belief acquisition was unconscious. -/- Bas van Fraassen's conception of a stance (cf. van Fraassen 2002)-a mix of commitments, values, goals-gives us reason to rethink the possibility of voluntarism. Might we be free to choose our stances for non-epistemic reasons? I argue here that the answer is negative. First, it is not clear at all whether stances can be objects of choice, properly speaking. And: Can we make sense of the idea of a free choice here? Furthermore, I argue that similar problems of Moore-paradoxality would arise for voluntarism about stances. If one does not want to completely devalue the idea of a stance, one should avoid voluntarism. (shrink)

The concept of empiricism evokes both a historical tradition and a set of philosophical theses. The theses are usually understood to have been developed by Locke, Berkeley, and Hume. But these figures did not use the term “empiricism,” and they did not see themselves as united by a shared epistemology into one school of thought. My dissertation analyzes the debate that elevated the concept of empiricism (and of an empiricist tradition) to prominence in English-language philosophy. -/- In the 1870s and (...) ’80s a lively debate about psychology emerged. Neo-Kantian idealists criticized the very idea that the mind can be studied scientifically. A group of philosopher-psychologists responded, often in Mind. They were among the first to call themselves “empiricists,” arguing that psychology could provide a scientific basis for philosophical progress. Idealists held that empirical psychology depended on premises developed by Locke, Berkeley, and Hume. These premises were allegedly absurd because they rendered ideas of extension, as well as other ideas crucial to natural science, unreal. Those who wanted to advance psychology towards becoming a legitimate science were forced to engage these philosophical attacks, while at the same time to develop empirical theories that could successfully explain some characteristics of experience. I show how James’s theory of space perception accomplished both tasks. -/- In developing this theory, James found he had to reject the Lockean notion that reality is associated with passively-registered sensations. James also abandoned Berkeley and Hume’s claim that ideas are ultimately derived from atomic sensations. Instead, James presented experimental evidence that sensation is a continuous stream. The mind must actively parse this stream if it is to gain a coherent representation of its environment. I argue that James’s stream-of-thought thesis served as a presupposition of his entire psychology. The thesis showed how the labor of investigating the mind could be divided between philosophers and scientists, and in a manner sensitive to the concerns of both. The stream thesis also provided a scientific basis for a new philosophical empiricism that, I argue, has a hidden legacy in the history of analytic philosophy. (shrink)