It is often presumed that the laws of nature have special significance for scientific reasoning. But the laws' distinctive roles have proven notoriously difficult to identify--leading some philosophers to question if they hold such roles at all. This study offers original accounts of the roles that natural laws play in connection with counterfactual conditionals, inductive projections, and scientific explanations, and of what the laws must be in order for them to be capable of playing these roles. Particular attention is given (...) to laws of special sciences, levels of scientific explanation, natural kinds, ceteris-paribus clauses, and physically necessary non-laws. (shrink)

Certain scientific explanations of physical facts have recently been characterized as distinctively mathematical –that is, as mathematical in a different way from ordinary explanations that employ mathematics. This article identifies what it is that makes some scientific explanations distinctively mathematical and how such explanations work. These explanations are non-causal, but this does not mean that they fail to cite the explanandum’s causes, that they abstract away from detailed causal histories, or that they cite no natural laws. Rather, in these explanations, (...) the facts doing the explaining are modally stronger than ordinary causal laws or are understood in the why question’s context to be constitutive of the physical arrangement at issue. A distinctively mathematical explanation works by showing the explanandum to be more necessary than ordinary causal laws could render it. Distinctively mathematical explanations thus supply a kind of understanding that causal explanations cannot. 1 Introduction2 Some Distinctively Mathematical Scientific Explanations3 Are Distinctively Mathematical Explanations Set Apart by their Failure to Cite Causes? 4 Distinctively Mathematical Explanations do not Exploit Causal Powers5 How these Distinctively Mathematical Explanations Work6 Conclusion. (shrink)

Many philosophers argue that explanatoriness plays no special role in confirmation – that “inference to the best explanation” (IBE) incorrectly demands giving hypotheses extra credit for their potential explanatory qualities beyond the credit they already deserve for their predictive successes. This paper argues against one common strategy for responding to this thought – that is, for trying to fit IBE within a Bayesian framework. That strategy argues that a hypothesis’ explanatory quality (its “loveliness”) contributes either to its prior probability or (...) to its likelihood. This paper argues that this strategy fails because it must give different treatments to two hypotheses that are unlovely for the very same reason. The strategy therefore loses the insight into scientific reasoning that its reconstruction in terms of IBE is supposed to provide. The paper then provides a Bayesian account of the confirmatory role of explanatoriness that represents explanatory quality as having an impact in the same place for two hypotheses that are unlovely for the very same reason. This approach works by “putting explanation back into IBE” – that is, by invoking hypotheses that refer explicitly to scientific explanation and by invoking the agent's background opinions regarding the kind of explanation that the evidence is liable to have. On this approach, there is no list of “explanatory virtues” the possession of which always helps to make an explanation better. Rather, for different facts, there are different characteristics that our background knowledge of other explanations gives us some reason to expect the given fact's explanation to possess. (shrink)

Ceteris-paribus clauses are nothing to worry about; aceteris-paribus qualifier is not poisonously indeterminate in meaning. Ceteris-paribus laws teach us that a law need not be associated straightforwardly with a regularity in the manner demanded by regularity analyses of law and analyses of laws as relations among universals. This lesson enables us to understand the sense in which the laws of nature would have been no different under various counterfactual suppositions — a feature even of those laws that involve no ceteris-paribus (...) qualification and are actually associated with exceptionless regularities. Ceteris-paribus generalizations of an‘inexact science’ qualify as laws of that science in virtue of their distinctive relation to counterfactuals: they form a set that is stable for the purposes of that field. (Though an accident may possess tremendous resilience under counterfactual suppositions, the laws are sharply distinguished from the accidents in that the laws are collectively as resilient as they could logically possibly be.) The stability of an inexact science's laws may involve their remaining reliable even under certain counterfactual suppositions violating fundamental laws of physics. The ceteris-paribus laws of an inexact science may thus possess a kind of necessity lacking in the fundamental laws of physics. A nomological explanation supplied by an inexact science would then be irreducible to an explanation of the same phenomenon at the level of fundamental physics. Island biogeography is used to illustrate how a special science could be autonomous in this manner. (shrink)

Unlike explanation in science, explanation in mathematics has received relatively scant attention from philosophers. Whereas there are canonical examples of scientific explanations, there are few examples that have become widely accepted as exhibiting the distinction between mathematical proofs that explain why some mathematical theorem holds and proofs that merely prove that the theorem holds without revealing the reason why it holds. This essay offers some examples of proofs that mathematicians have considered explanatory, and it argues that these examples suggest a (...) particular account of explanation in mathematics. The essay compares its account to Steiner's and Kitcher's. Among the topics that arise are proofs that exploit symmetries, mathematical coincidences, brute-force proofs, simplicity in mathematics, merely clever proofs, and proofs that unify what other proofs treat as separate cases. (shrink)

Hempel and Giere contend that the existence of provisos poses grave difficulties for any regularity account of physical law. However, Hempel and Giere rely upon a mistaken conception of the way in which statements acquire their content. By correcting this mistake, I remove the problem Hempel and Giere identify but reveal a different problem that provisos pose for a regularity account — indeed, for any account of physical law according to which the state of affairs described by a law-statement presupposes (...) a Humean regularity. These considerations suggest a normative analysis of law-statements. On this view, law-statements are not distinguished from accidental generalizations by the kind of Humean regularities they describe because a law-statement need not describe any Humean regularity. Rather, a law-statement says that in certain contexts, one ought to regard the assertion of a given type of claim, if made with justification, as a proper way to justify a claim of a certain other kind. (shrink)

Among the niftiest arguments for scientific anti-realism is the ‘pessimistic induction’ (also sometimes called ‘the disastrous historical meta-induction’). Although various versions of this argument differ in their details (see, for example, Poincare 1952: 160, Putnam 1978: 25, and Laudan 1981), the argument generally begins by recalling the many scientific theories that posit unobservable entities and that at one time or another were widely accepted. The anti-realist then argues that when these old theories were accepted, the evidence for them was quite (...) persuasive – roughly as compelling as our current evidence is for our best scientific theories positing various unobservable entities. Nevertheless, the anti-realist argues, most of these old theories turned out to be incorrect in the unobservables they posited. Therefore, the anti-realist concludes that with regard to the theories we currently accept, we should believe that probably, most of them are likewise incorrect in the unobservable entities they posit. (This argument appeals to what our best current theories say about unobservables in order to show that the entities posited by some earlier theory are not real. So the argument takes the form of a reductio of the view that the apparent success of some scientific theory justifies our believing in its accuracy regarding unobservables.) Of course, this argument has been criticized on many grounds. Some have argued, for instance, that the scientific theories we currently accept are much better supported than were earlier scientific theories at the time they were accepted. In addition, some have argued that many scientific theories accepted justly in the past were in fact accurate.. (shrink)

Several foundational documents of bioethics mention the special obligation researchers have to vulnerable research participants. However, the treatment of vulnerability offered by these documents often relies on enumeration of vulnerable groups rather than an analysis of the features that make such groups vulnerable. Recent attempts in the scholarly literature to lend philosophical weight to the concept of vulnerability are offered by Luna and Hurst. Luna suggests that vulnerability is irreducibly contextual and that Institutional Review Boards (Research Ethics Committees) can only (...) identify vulnerable participants by carefully examining the details of the proposed research. Hurst, in contrast, defines the vulnerable as those especially at risk of incurring the wrongs to which all research ethics participants are exposed. We offer a more substantive conception of vulnerability than Luna but one that gives rise to a different rubric of responsibilities from Hurst's. While we understand vulnerability to be an ontological condition of human existence, in the context of research ethics, we take the vulnerable to be research subjects who are especially prone to harm or exploitation. Our analysis rests on developing a typology of sources of vulnerability and showing how distinct sources generate distinct obligations on the part of the researcher. Our account emphasizes that the researcher's first obligation is not to make the research participant even more vulnerable than they already are. To illustrate our framework, we consider two cases: that of a vulnerable population involved in international research and that of a domestic population of people with diminished capacity. (shrink)

Really statistical explanation is a hitherto neglected form of noncausal scientific explanation. Explanations in population biology that appeal to drift are RS explanations. An RS explanation supplies a kind of understanding that a causal explanation of the same result cannot supply. Roughly speaking, an RS explanation shows the result to be mere statistical fallout.

Batterman and Rice offer an account of “minimal model explanations” and argue against “common features accounts” of those explanations. This paper offers some objections to their proposals and arguments. It argues that their proposal cannot account for the apparent explanatory asymmetry of minimal model explanations. It argues that their account threatens ultimately to collapse into a “common features account.” Finally, it argues against their motivation for thinking that an explanation appealing to “common features” would have to explain the common features’ (...) own prevalence. (shrink)

Fine and Rosen have argued that normative necessity is distinct from and weaker than metaphysical necessity. The first aim of this paper is to specify what it would take for this view to be true—that is, what normative necessity would have to be like. The author argues that in order for normative necessity to be weaker than metaphysical necessity, the metaphysical necessities must all be preserved under every counterfactual antecedent with which they are all collectively logically consistent—even when their preservation (...) requires that a normative necessity fail to be preserved. By exhibiting some examples that fail to display this pattern of counterfactual invariance, the author argues against the view that normative necessity is weaker than metaphysical necessity. To give this argument is the second aim of this paper. (shrink)

Many philosophers have believed that the laws of nature differ from the accidental truths in their invariance under counterfactual perturbations. Roughly speaking, the laws would still have held had q been the case, for any q that is consistent with the laws. (Trivially, no accident would still have held under every such counterfactual supposition.) The main problem with this slogan (even if it is true) is that it uses the laws themselves to delimit qs range. I present a means of (...) distinguishing the laws (and their logical consequences) from the accidents, in terms of their range of invariance under counterfactual antecedents, that does not appeal to physical modalities (or any cognate notion) in delimiting the relevant range of counterfactual perturbations. I then argue that this approach explicates the sense in which the laws possess a kind of necessity. (shrink)

Explanation in mathematics has recently attracted increased attention from philosophers. The central issue is taken to be how to distinguish between two types of mathematical proofs: those that explain why what they prove is true and those that merely prove theorems without explaining why they are true. This way of framing the issue neglects the possibility of mathematical explanations that are not proofs at all. This paper addresses what it would take for a non-proof to explain. The paper focuses on (...) a particular example of an explanatory non-proof: an argument that mathematicians regard as explaining why a given theorem holds regarding the derivative of an infinite sum of differentiable functions. The paper contrasts this explanatory non-proof with various non-explanatory proofs of the same theorem. The paper offers an account of what makes the given non-proof explanatory. This account is motivated by investigating the difficulties that arise when we try to extend Mark Steiner’s influential account of explanatory proofs to cover this explanatory non-proof. (shrink)

A conservation law in physics can be either a constraint on the kinds of interaction there could be or a coincidence of the kinds of interactions there actually are. This is an important, unjustly neglected distinction. Only if a conservation law constrains the possible kinds of interaction can a derivation from it constitute a scientific explanation despite failing to describe the causal/mechanical details behind the result derived. This conception of the relation between “bottom-up” scientific explanations and one kind of “top-down” (...) scientific explanation is motivated by several examples from classical and modern physics. (shrink)

This paper proposes an account of why proof is the only way to acquire knowledge of some mathematical proposition’s truth. Admittedly, non-deductive arguments for mathematical propositions can be strong and play important roles in mathematics. But this paper proposes a necessary condition for knowledge that can be satisfied by putative proofs (and proof sketches), as well as by non-deductive arguments in science, but not by non-deductive arguments from mathematical evidence. The necessary condition concerns whether we can justly expect that if (...) the mathematical proposition is actually false, despite the evidence for its truth, then this fact has an explanation. (shrink)

Scientific essentialism aims to account for the natural laws' special capacity to support counterfactuals. I argue that scientific essentialism can do so only by resorting to devices that are just as ad hoc as those that essentialists accuse Humean regularity theories of employing. I conclude by offering an account of the laws' distinctive relation to counterfactuals that portrays laws as contingent but nevertheless distinct from accidents by virtue of possessing a genuine variety of necessity.

This paper examines some recent attempts that use counterfactuals to understand the asymmetry of non-causal scientific explanations. These attempts recognize that even when there is explanatory asymmetry, there may be symmetry in counterfactual dependence. Therefore, something more than mere counterfactual dependence is needed to account for explanatory asymmetry. Whether that further ingredient, even if applicable to causal explanation, can fit non-causal explanation is the challenge that explanatory asymmetry poses for counterfactual accounts of non-causal explanation. This paper argues that several recent (...) accounts Explanation beyond causation: philosophical perspectives on non-causal explanations, Oxford University Press, Oxford, pp 117–140, 2018; Jansson and Saatsi in Br J Philos Sci, forthcoming; Jansson in J Philos 112:7–599, 2015; Saatsi and Pexton in Philos Sci 80: 613–624, 2013; French and Saatsi, in: Reutlinger and Saatsi Explanation beyond causation: philosophical perspectives on non-causal explanations, Oxford University Press, Oxford, pp 185–205, 2018) fail to meet this challenge. The paper then sketches a more positive proposal for dealing with explanatory asymmetry in non-causal explanations. (shrink)

This paper argues that in at least some cases, one proof of a given theorem is deeper than another by virtue of supplying a deeper explanation of the theorem — that is, a deeper account of why the theorem holds. There are cases of scientific depth that also involve a common abstract structure explaining a similarity between two otherwise unrelated phenomena, making their similarity no coincidence and purchasing depth by answering why questions that separate, dissimilar explanations of the two phenomena (...) cannot correctly answer. The connections between explanation, depth, unification, power, and coincidence in mathematics and science are compared. (shrink)

Several philosophers have suggested that some scientific explanations work not by virtue of describing aspects of the world’s causal history and relations, but rather by citing mathematical facts. This paper investigates what mathematical facts could be in order for them to figure in such “distinctively mathematical” scientific explanations. For “distinctively mathematical explanations” to be explanations by constraint, mathematical language cannot operate in science as representationalism or platonism describes. It can operate as Aristotelian realism describes. That is because Aristotelian realism enables (...) mathematics to apply to the physical world without a morphism’s mediation. (shrink)

Some mathematical proofs explain why the theorems they prove hold. This paper identifies several challenges for any counterfactual account of explanation in mathematics (that is, any account according to which an explanatory proof reveals how the explanandum would have been different, had facts in the explanans been different). The paper presumes that countermathematicals can be nontrivial. It argues that nevertheless, a counterfactual account portrays explanatory power as too easy to achieve, does not capture explanatory asymmetry, and fails to specify why (...) certain proofs are explanatory and others are not. Greater informativeness about counterfactual dependence can even yield less explanatory power. (shrink)

Rosenberg has recently argued that explanations supplied by (what he calls) functional biology are mere promissory notes for macromolecular adaptive explanations. Rosenberg's arguments currently constitute one of the most substantial challenges to the autonomy, irreducibility, and indispensability of the explanations supplied by functional biology. My responses to Rosenberg's arguments will generate a novel account of the autonomy of functional biology. This account will turn on the relations between counterfactuals, scientific explanations, and natural laws. Crucially, in their treatment of the laws' (...) relation to counterfactuals, Rosenberg's arguments beg the question against the autonomy of functional biology. This relation is considerably more subtle than is suggested by familiar slogans such as Laws support counterfactuals; accidents don't. (shrink)

Why do forces compose according to the parallelogram of forces? This question has been controversial; it is one episode in a longstanding, fundamental dispute regarding which facts are not to be explained dynamically. If the parallelogram law is explained statically, then the laws of statics are separate from and “transcend” the laws of dynamics. Alternatively, if the parallelogram law is explained dynamically, then statical laws become mere corollaries to the dynamical laws. I shall attempt to trace the history of this (...) controversy in order to identify what it would be for one or the other of these rival views to be correct. I shall argue that various familiar accounts of natural law not only make it difficult to see what the point of this dispute could have been, but also improperly foreclose some serious scientific options. I will sketch an alternative account of laws that helps us to understand what this dispute was all about. (shrink)

This paper analyzes the logical truths as (very roughly) those truths that would still have been true under a certain range of counterfactual perturbations.What’s nice is that the relevant range is characterized without relying (overtly, at least) upon the notion of logical truth. This approach suggests a conception of necessity that explains what the different varieties of necessity (logical, physical, etc.) have in common, in virtue of which they are all varieties of necessity. However, this approach places the counterfactual conditionals (...) in an unfamiliar foundational role. (shrink)

Sober 2011 argues that, contrary to Hume, some causal statements can be known a priori to be true—notably, some ‘would promote’ statements figuring in causal models of natural selection. We find Sober's argument unconvincing. We regard the Humean thesis as denying that causal explanations contain any a priori knowable statements specifying certain features of events to be causally relevant. We argue that not every ‘would promote’ statement is genuinely causal, and we suggest that Sober has not shown that his examples (...) of ‘would promote’ statements manage to achieve a priori status without sacrificing their causal character. (shrink)

. A typical textbook explanation of a rocket’s motion when its engine is fired appeals to momentum conservation: the rocket accelerates forward because its exhaust accelerates rearward and the system’s momentum must be conserved. This paper examines how this explanation works, considering three challenges it faces. First, the explanation does not proceed by describing the forces causing the rocket’s motion. Second, the rocket’s motion has a causal-mechanical explanation involving those forces. Third, if momentum conservation and the rearward motion of the (...) rocket’s exhaust explain why the rocket accelerates forward, then presumably momentum conservation and the rocket’s forward motion likewise explain why the rocket emits exhaust rearward. Explanatory circularity threatens to follow from this pair of explanations. This paper examines how the conservation-law explanation works and how it is compatible with the causal-mechanical explanation. The paper argues that these two explanations do not explain precisely the same fact relative to the same contrast class. The paper interprets the two conservation-law explanations as non-causal and argues that they yield no explanatory circularity. (shrink)

In a recent paper replying to the inductive sceptic, Samir Okasha says that the Humean argument for inductive scepticism depends on mistakenly construing inductive reasoning as based on a principle of the uniformity of nature. I dispute Okasha's argument that we are entitled to the background beliefs on which (he says) inductive reasoning depends. Furthermore, I argue that the sorts of theoretically impoverished contexts to which a uniformity-of-nature principle has traditionally been restricted are exactly the contexts relevant to the inductive (...) sceptic's argument, and (pace Okasha) are not at all remote from actual scientific practice. I discuss several scientific examples involving such contexts. (shrink)

Roche and Sober have offered a new argument for the view that a hypothesis H is not confirmed by its capacity to explain some observation O. Their argument purports to work by showing that O screens H off from the fact that H would explain O. This paper offers several objections to this argument. Firstly, the screening-off test cannot identify whatever evidential contribution Hs explanatoriness may make. Secondly, that H would explain O may be logically necessary, eluding the screening-off test. (...) Thirdly, the test cannot detect an important difference that Hs explanatoriness often makes to Hs confirmation. (shrink)

Roche and Sober (2013; 2014; 2017; 2019) have offered an important new argument that explanatoriness lacks confirmatory significance. My aim in this paper is not only to contend that their argument fails to show that in confirmation ‘there is nothing special about explanatoriness’ (Roche & Sober, 2017: 589), but also to reveal what is special confirmationwise about explanatoriness. I will argue that much of the heavy work in Roche and Sober's argument is done by the dichotomy into which they carve (...) up the philosophical options: either explanatory considerations per se are confirmatorily relevant or they acquire whatever confirmatory significance they may possess only by way of background beliefs and so no differently from non‐explanatory considerations. I will argue that this is a false dichotomy: Explanatoriness per se can have confirmatory relevance even while explanatory considerations acquire their confirmatory relevance only through background opinions. Furthermore, there is ‘something special’ about the background opinions concerning explanations that allow explanatoriness in itself to have confirmatory impact. Unlike other sorts of background, having some background opinions expressly about explanations is indispensable to being an observer at all. (shrink)

A long lasting debate in the field of implicit learning is whether participants can learn without acquiring conscious knowledge. One crucial problem is that no clear criterion exists allowing to identify participants who possess explicit knowledge. Here, we propose a method to diagnose during a serial reaction time task those participants who acquire conscious knowledge. We first validated this method by using Stroop-like material during training. Then we assessed participants’ knowledge with the Inclusion/Exclusion task and the wagering task . Both (...) experiments confirmed that for participants diagnosed as having acquired conscious knowledge about the underlying sequence the Stroop congruency effect disappeared, whereas for participants not diagnosed as possessing conscious knowledge it only slightly decreased. In addition, both experiments revealed that only participants diagnosed as conscious were able to strategically use their acquired knowledge. Thus, our method allows to reliably distinguish between participants with and without conscious knowledge. (shrink)

Precise measurements of well-being would be of profound societal importance. Yet, the sceptical worry that we cannot use social science instruments and tests to measure well-being is widely discussed by philosophers and scientists. A recent and interesting philosophical argument has pointed to the psychometric procedures of construct validation to address this sceptical worry. The argument has proposed that these procedures could warrant confidence in our ability to measure well-being. The present paper evaluates whether this type of argument succeeds. The answer (...) is that it depends on which methodological background assumptions are motivating the sceptical worry to begin with. We show this by doing two things. First, we clarify (a) the different types of well-being theories involved in the science of well-being, and (b) the general methodological dimensions of well-being theorising. Second, we apply these distinctions and argue that construct validation is an unsuccessful response to measurement scepticism if this scepticism is motivated by a form of methodological non-naturalism. In the light of this, the overall point of the paper is that philosophers and scientists, when discussing measurement of well-being, should explicate their deeper methodological commitments. We further suggest that making such explicit commitments might present philosophers with a dilemma. (shrink)

Myrvold (2003) has proposed an attractive Bayesian account of why theories that unify phenomena tend to derive greater epistemic support from those phenomena than do theories that fail to unify them. It is argued, however, that "unification" in Myrvold's sense is both too easy and too difficult for theories to achieve. Myrvold's account fails to capture what it is that makes unification sometimes count in a theory's favor.

Why should science be so interested in discovering whether p is a law over and above whether p is true? The answer may involve the laws' relation to counterfactuals: p is a law iff p would still have obtained under any counterfactual supposition that is consistent with the laws. But unless we already understand why science is especially concerned with the laws, we cannot explain why science is especially interested in what would have happened under those counterfactual suppositions consistent with (...) the laws. It is argued that the laws form the only non-trivially "stable" set, where "stability" is invariance under a certain range of counterfactual suppositions not itself defined by reference to the laws. It is then explained why science should be so interested in identifying a non-trivially "stable" set: because of stability's relation to the best set of "inductive strategies". (shrink)

Very little has been done to find out what corporations have done to build ethical values into their organizations. In this report on a survey of 1984 Fortune 1000 industrial and service companies the Center for Business Ethics reveals some facts regarding codes of ethics, ethics committees, social audits, ethics training programs, boards of directors, and other areas where corporations might institutionalize ethics. Based on the survey, the Center for Business Ethics is convinced that corporations are beginning to take steps (...) to institutionalize ethics, while recognizing that in most cases more specific mechanisms and strategies need to be implemented to make their ethics efforts truly effective. (shrink)

The merits of David Lewis’s Best System Account of natural law are frequently debated. But to my knowledge, the prospects for extending the BSA to cover meta-laws have never been examined. I shall identify two obstacles facing the most natural way of extending the BSA to cover meta-laws. The BSA’s fans should consider how these obstacles are to be overcome. Meta-laws are laws about laws. For example, Einstein’s special theory of relativity incorporates a meta-law: The content of the [special] relativity (...) theory can … be summarized in one sentence: all natural laws must be so conditioned that they are covariant with respect to Lorentz transformations. [The special theory of relativity] is not a theory in the usual sense but is better regarded as a second-level theory, or a theory of theories that constrains first-level theories. The principle of relativity is an example of a symmetry principle: a principle requiring that the first-order laws be unchanged under a given transformation. Long before Einstein proposed the principle of relativity, other spacetime symmetries were widely believed to be meta-laws: that the first-order laws are covariant under arbitrary spatial displacements, temporal displacements and spatial rotations. These spacetime symmetries require the laws to treat all spatial locations and directions alike and all moments alike. For instance, symmetry under temporal displacements rules out a fundamental force law specifying that a given force declines with the inverse-square of the distance before a given moment but with the inverse-cube of the distance at and after that moment. Wigner characterizes such a symmetry principle as ‘a superprinciple which is in …. (shrink)

Scientific realism is the view that one can be justified in believing, of some theory about unobservable entities, that the entities it posits are real and accurately described by the theory, in the same sense as one can be justified in believing that the theory’s empirical predictions are accurate, and that so to believe is what it means for a scientist to “accept” that theory, because the goal of science is to describe reality, even its unobservable features. The first part (...) is an epistemic claim, whereas the second concerns the rational reconstruction of scientific practice. Together they suggest that evidence justifying scientific acceptance of a theory justifies belief that the unobservable entities it posits are real and as the theory purports them to be. Anti-realism denies ; it holds that claims about unobservables are either unjustified or justified in a different sense from predictions of observations. It also denies ; it maintains that in her scientific work, as rationally reconstructed, a scientist who accepts a claim about unobservables believes that it yields accurate predictions of observables, not that it describes unobservables. (shrink)

With the rise and public accessibility of AI-enabled decision-support systems, individuals outsource increasingly more of their decisions, even those that carry ethical dimensions. Considering this trend, scholars have highlighted that uncritical deference to these systems would be problematic and consequently called for investigations of the impact of pertinent technology on humans’ ethical decision-making. To this end, this article conducts a systematic review of existing scholarship and derives an integrated framework that demonstrates how intelligent decision-support systems (IDSSs) shape humans’ ethical decision-making. (...) In particular, we identify resulting consequences on an individual level (i.e., deliberation enhancement, motivation enhancement, autonomy enhancement and action enhancement) and on a societal level (i.e., moral deskilling, restricted moral progress and moral responsibility gaps). We carve out two distinct methods/operation types (i.e., processoriented and outcome-oriented navigation) that decision-support systems can deploy and postulate that these determine to what extent the previously stated consequences materialize. Overall, this study holds important theoretical and practical implications by establishing clarity in the conceptions, underlying mechanisms and (directions of) influences that can be expected when using particular IDSSs for ethical decisions. (shrink)