Mario Alai

The empirical underdetermination of theories is a philosophical problem which until the last century has not seriously troubled actual science. The reason is that confirmation does not depend only on empirical consequences, and theoretical virtues allow to choose among empirically equivalent theories. Moreover, I argue that the theories selected in this way are not just pragmatically or aesthetically better, but more probably (and/ or largely) true. At present in quantum mechanics not even theoretical virtues allow to choose among many competing theories and interpretations, but this is because none of them possess those virtues to a sufficient degree. However, first, we can hope for some future advancement (new empirical tests, or new theories). Second, even if no further progress came forth, all the most credited competitors agree on a substantial core of theoretical assumptions. Therefore underdetermination does not show that we cannot be realist on unobservable entities in general, but at most that in particular fields our inquiry may encounter some de facto limits.

Currently one of the most plausible versions of scientific realism is “Deployment” (or “Partial”, or “Conservative”) Realism, based on various contributions in the recent literature (especially Kitcher 1993), and worked out as a unitary account in Psillos (1999). According to this version we can believe that theories are at least partly true (because that is the best
explanation for their predictive success—especially novel predictions), and discarded theories which had novel predictive success had nonetheless some true parts: those necessary to derive their novel predictions. In fact, it has been argued that the partial truth of theories (including the discarded ones) is the only non-miraculous explanation of their success.
According to Doppelt (2005, 2007) this account cannot withstand the antirealist objections based on the “pessimistic meta-induction” and Laudan’s historical counterexamples. Moreover, it is incomplete, as it purports to explain the predictive success of theories, but overlooks the necessity of also explaining their explanatory success. Accordingly, he
proposes a new version of realism, presented as the best explanation for both predictive and explanatory success, and committed only to the truth of the best current theories, not that of the discarded ones (Doppelt 2007, 2011, 2013, 2014).
I have argued elsewhere (Alai 2016a) that Doppelt’s “Best Theory Realism” is not really a viable option, for it can explain neither the success of past theories nor their failures; moreover, it is rather implausible, and actually the easiest prey of the pessimistic meta-induction argument. Here instead I argue for the following claims: (a) Doppelt has not shown that Deployment Realism as it stands cannot solve the problems raised by the
history of science; (b) explaining explanatory success does not add much to the explanation of novel predictive success; (c) Doppelt is right that truth is not a sufficient explanans, but for different reasons than he thinks, and this does not refute Deployment Realism, but helps to detail it better. In a more explicit formulation, the realist Inference to the Best Explanation (IBE) concludes not only that theories are true, but also that the scientific method and scientists are reliable, nature is orderly and simple, and background
theories are approximately true.

Agazzi stages a complete, very detailed and overall convincing defence of scientific realism, its presuppositions and corollaries (mind-independence of reality, referentiality of theories, truth as correspondence, knowledge as the goal of science, justifiability of beliefs in unobservables through abductive arguments). But his claims that truth is relative to a circumscribed domain, not " pictorial " and not pertaining to theories, and that scientific objects certainly exist because they are nothing but abstract bundles of properties, are potentially ambiguous. Moreover, to the antirealist objections based on radical theory-change he replies that pre-and post-revolutionary theories do not contradict each other and are equally true, because each one deals with a different domain of objects of its own making: but this reply (apparently a legacy of neopositivistic operationalism) risks to make theories analytic, so slipping into conventionalism, or to reduce their content to observable phenomena, thus giving in to antirealism.

Il saggio analizza gli argomenti in favore e contro rispettivamente del dualismo semantico di Frege e del monismo semantico di Russell, cercando poi di mostrare come ciascuno dei due abbia ragione, rispettivamente in certi sensi ed entro certi limiti

Strong predictivism, the idea that novel predictions per se confirm theories more than accommodations, is based on a " no miracle " argument from novel predictions to the truth of theories (NMAT). Eric Barnes rejects both: he reconstructs the NMAT as seeking an explanation for the entailment relation between a theory and its novel consequences, and argues that it involves a fallacious application of Occam's razor. However, he accepts a no miracle argument for the truth of background beliefs (NMABB): scientists endorsed a successful theory because they were guided by largely true background beliefs. This in turn raises the probability that the theory is true; so Barnes embraces a form of weak predictivism, according to which predictions are only indirectly relevant to confirmation. To Barnes I reply that we should also explain how the successful theory constructed, not just endorsed; background beliefs are not enough to explain success, scientific method must also be considered; Barnes can account for some measure of confirmation of our theories, but not for the practical certainty conferred to them by some astonishing predictions; true background beliefs and reliability by themselves cannot explain novel success, the truth of theories is also required. Hence, the NMAT is sound, and strong predictivism is right. In fact, Barnes misinterprets the NMAT, which does not involve Occam's razor, takes as explanandum the building of a theory which turned out to predict surprising facts, and successfully concludes that the theory is true. This accounts for the practically certain confirmation of our most successful theories, in accordance with strong predictivism.

Evandro Agazzi's volume Scientific Objectivity and its Contexts is here introduced. First, the genesis and the content of the book are outlined. Secondly, an overview of Agazzi's philosophy of science is provided. Its main roots are epistemological realism in the Aristotelian/scholastic tradition, and contemporary science-oriented epistemology, especially in Logical Empiricism. As a result, Agazzi's thought is nicely balanced between empiricism and rationalism, it avoids gnoseologistic dualism by stressing the intentionality of knowledge, and it insists on the operational and referential character of science. Finally, an account is given of Agazzi's view of the origin and nature of scientific objects, which allows to understand how his sophisticated and " perspectival " realism differs both from naïve realism and constructivism.

Debates on realism in science concern two main questions: whether theoretical knowledge is possible, and whether it is objective. Today, as in the past, the possibility of theoretical knowledge is often denied because of the empirical underdetermination of theories. Realists rely on explanatory power, theoretical virtues, and instrumental but theory-free observation to solve this problem. Besides, they use the " no miracles argument " for the truth of successful theories. Antirealists, however, deny that explanation is either necessary or possible, and that is a cue to truth. Moreover, they reject realism and the cogency of the no miracle argument by the pessimistic induction from the falsity of since past successful theories. Some realists reply that there is a radical discontinuity between past science (largely off-track) and current science (basically sound). But this reply is at best insufficient, and most realists prefer to restrict their commitment to selected parts or features of theories, both past and present. Forms of " selective realism " are entity realism, structural realism, deployment realism and semirealism, but also the verisimilitude research program and the restricted-domain approach. Realists need criteria to identify the true components of theories, and a noteworthy candidate is essential involvement in functionally novel and surprising predictions. The second main question is a special instance of the old debate between realists and relativists or idealists: according to antirealists science cannot be objective, because of its inherently " perspectival " nature, characterized by a priori and subjective factors. On the contrary, perspectival realists argue the specific " viewpoints " within which scientists must work do not prevent them to discover objective features of reality.

Umberto Eco wonders how Ferraris' New Realism differs from the Old Realism, of which it shares the correspondence view of truth and knowledge; so he suggests in its place a " negative " realism. Paolo Parrini challenges both Ferraris and Eco to produce arguments showing that the unamendable object of our knowledge is a metaphysically independent reality, rather than simply experience itself; and since he believes that this cannot be shown, he proposes instead an " empirical " realism. I argue that Ferraris' realism is far from unsophisticated, and Eco himself must either grant some form of correspondence, or fall back into post-modern relativism. Further, I suggest that a merely " empirical " realism does not fully satisfy our demand for knowledge, and I sketch strategies for answering Parrini's challenge. For instance, who claims that we cannot know reality in itself swaps the usual sense of 'knowing' with a new and unworkable one.

1. Two Pessimistic Arguments against Realism and Two Defence Strategies We can broadly understand Scientific Realism (ScR) (as opposed to van Fraassen's constructive empiricism) as the claim that there can be compelling reasons to believe in the at least partial and/or approximate truth of scientific theories. ScR is notoriously supported by the " no miracle " argument (NMA): NMA (α) some theories are successful; (β) the best or only explanation of success is truth, therefore (γ) successful theories are most probably true; therefore (δ) there can be compelling reason to believe that theories are true (ScR) But there are two important arguments against realism from the failures of past science. They move from a similar premise and are strictly connected, so that many confuse them under the name of " pessimistic meta-induction " ;1 but they are distinct and have different conclusions. The former, which more properly deserves that name, is: PMI (I) most past successful theories were radically false; (II) there is no radical (epistemic, methodological, etc.) difference between past and present theories; therefore, by induction (III) also current and future successful theories are and will most likely be radically false; therefore (IV) there is no compelling reason to believe that theories are true (in fact, there are reasons to believe that they are mostly radically false). PMI was famously formulated by Poincaré (1902, 160) and Putnam (1978, 25). It has the typical form of induction, and its conclusion directly contradicts ScR. The latter antirealistic argument from past failures of science is MMT (A) many past successful theories were radically false (= not true); therefore (B) truth is not the best or only explanation of success; therefore (C) The NMA does not offer compelling reason to believe that theories are true. This argument has been proposed by Laudan (1981) and in a particularly cogent form by Lyons (2002). Its conclusion does not contradict ScR, but only premise (β) and so the cogency of the NMA. Thus, if it succeeds we lose one argument for ScR, but ScR might still be true. Moreover, it is not inductive, but deductive. Lyons (2002) called it " meta-modus tollens " (MMT), because in the NMA (β) entails (γ), and MMT rejects (β), via modus tollens by denying (γ). Realists have various replies to these attacks, but since PMI and MMT are different, not each reply is equally effective against both; besides, each reply works by qualifying the broad characterization of ScR in one way or another; so, its success does not depend only on how well it resists PMI and/or MMT, but also on how plausible is the form of realism it configures; finally, these replies can be subsumed under two broad strategies, the selective strategy and the discontinuity strategy:-the selective strategy claims that although the rejected but successful theories may have been partly false, they were not radically false, in the sense that they included some significant true assumptions or components. So it rejects the first premise of PMI and MMT, the radical falsity of past theories. Hence, PMI can conclude at most that theories are partly false. As a result, selective realists stress the continuity between 1 For instance, Doppelt calls both of them 'pessimistic meta-induction'; Saatsi distinguishes them, but curiously he calls the MMT 'pessimistic meta-induction''

Kurosawa's intriguing masterpiece Rashōmon might appear as a manifesto of post-modern relativism, according to which we cannot know any " objective reality " , but only alternative subjective " versions " of the world. This is also what some quotes by Kurosawa seem to suggest. No doubt, the accounts of an event given from alternative conceptual, epistemic, or psychological perspectives may be very different. Yet, such stories may all correspond to the actual world, and often, by relativising them to the subjects' standpoints, we can grasp some important features of the subject-independent reality. In fact, by this method the actual course of events in Rashōmon can be reconstructed in a fairly plausible and unambiguous way. Although the spectator can gather her clues only from the apparently confusing and unreliable accounts of the various characters, perhaps Kurosawa's story is less enigmatic than he himself thought. The script also bears striking resemblances to the puzzles of quantum mechanics. Actually, if current theories are right, microphysical reality is even more elusive than historic reality, portrayed by Kurosawa through the deforming mirrors of subjective drives and motivations. But upon reflection, even in quantum mechanics there is no need to give up the idea that there exists an objective, if awkward, reality, and that we can know it at least partially and perspectively.

Criticisms à la Laudan can block the “no miracles” argument for the (approximate) truth of whole theories. Realists have thus retrenched, arguing that at least the individual claims deployed in the derivation of novel predictions should be considered (approximately) true. But for Lyons (2002) there are historical counterexamples even to this weaker “deployment” realism: he lists a number of novel predictions supposedly derived from (radically) false claims. But if so, those successes would seem unexplainable, even by Lyons’ “modest surrealism” or other surrogates to realism. In fact, I argue, some of those predictions were an easy guess, or independently probable in the light of available evidence; hence, they are no counterexamples to deployment realism, for the no miracles argument wouldn’t apply to them.  In other instances, pace Lyon, the prediction was actually false, and could be reinterpreted as true only by interpreting as true also the claim from which it was derived; or again, a false claim was employed in the derivation of a true prediction, but inessentially, the essential role being played by a weaker true claim. But as soon as the paradoxical air of such historical cases is explained away in any of these manners, they cease to represent counterexamples to deployment realism.  If, as I suggest, all of them can be dealt with by these strategies, a theoretical claim can still be assumed to be true if it is crucial in deriving an improbable novel prediction.

This is an attempt to sort out what is it that makes many of us uncomfortable with the perdurantist solution to the problem of change. Lewis argues that only perdurantism can reconcile change with persistence over time, while neither presentism nor endurantism can. So, first, I defend the endurantist solution to the problem of change, by arguing that what is relative to time are not properties, but their possession. Second, I explore the anti-perdurantist strategy of arguing that Lewis cannot solve the problem of change, for he cannot account for how some properties are possessed by objects in time. However, I argue that this strategy fails, for if by saying that objects in time can have properties ‘timelessly’ we mean ‘‘at no particular time’’ and ‘‘tenselessly’’, only objects outside time can have properties in
that way; but if we mean ‘‘for all the time they exist’’, or ‘‘essentially’’, perdurantists can account for this. Finally, I argue that actually perdurantism cannot solve the problem, but for different reasons: for either it sweeps the problem under the carpet, denying change, and in general subverting our conceptual scheme in a dangerous way, or it becomes equivalent to the endurantist picture that properties are had at
times. Nor perdurantism is justified by the Relativity Theory or the B-theory of time, because while endurantism is certainly comfortable with presentism, it need not be committed to it; and even if it were, presentism need not be refuted by the Relativity Theory.

There are two possible realist defense strategies against the pessimistic
meta-induction and Laudan’s meta-modus tollens: the selective strategy, claiming that discarded theories are partially true, and the discontinuity strategy, denying that pessimism about past theories can be extended to current ones.Aradical version of discontinuity realism is proposed by Gerald Doppelt: rather than discriminating between true and false components within theories, he holds that superseded theories cannot
be shown to be even partially true (except insofar they agree with current ones), while present best theories are demonstrably completely true. I argue that this position, running counter both the cumulativity of science and fallibilism, is untenable; it cannot account for the success of past theories, nor for the failures of current theories, and rather than shutting the door to the pessimistic historical objections it opens it wide.
The best strategy, instead, joins the selective idea there was both some truth and some falsity in discarded theories, like in current ones, with the moderate discontinuity idea that the truth rate in present best theories is much greater than in past ones.

In 1936, after the “liberalization of empiricism”, Schlick and Carnap thought they could accept C.I. Lewis’ claim “If all minds disappeared from the universe, the stars would still go on in their courses” as a scientific truth, without accepting the metaphysical thesis of the mind independence of the external world. But this compromise failed, as it was based on mistaken antirealist semantic views: they accepted Lewis’ sentence, not the proposition it expresses, or its consequences. This is not to say that science is enough to support metaphysical realism, since it takes philosophy to show where they went wrong.

The currently most plausible version of scientific realism is probably “deployment” (or “partial”, or “conservative”) realism, based on various contributions in the recent literature, and worked out as a unitary account in Psillos (1999). According to it we can believe in the at least partial truth of theories, because that is the best (or the only non-miraculous) explanation of their predictive success (especially novel predictive success), and discarded theories which had novel predictive success had nonetheless some true parts, those necessary to derive their novel predictions.
According to Doppelt (2005, 2007) this account cannot withstand the antirealist objections based on the “pessimistic meta-induction” and Laudan’s historical counterexamples. Moreover it is incomplete, as it purports to explain the predictive success of theories, but overlooks the necessity to explain also their explanatory success. Accordingly, he proposes a new version of realism, presented as the best explanation of both predictive and explanatory success, and committed only to the truth of best current theories, not of the discarded ones.
Here I argue that (a) Doppelt has not shown that deployment realism as it stands cannot solve the problems raised by the history of science, (b) explaining explanatory success does not add much to explaining novel predictive success, and (c) a realism confined to current theories is implausible, and actually the easiest prey to the pessimistic meta-induction argument.

To begin with I distinguish various kinds of realism, especially commonsense, scientific and metaphysical realism. I then argue that all of them can be supported by explanationist arguments, among which I distinguish abduction, inference to the best explanation, and various forms of " no miracle " argument. Some of these arguments are based on first level explananda (empirical regularities and scientific phenomena), others on meta-level explananda (the success of science). Some are scientific inferences, some are philosophical. The " no miracle argument " is a meta-level philosophical argument, but some of its forms can be derived by equivalent transformations from first-level scientific arguments, so naturalists claimed that it is a scientific argument. While granting its scientific bases, however, I argue that it is distinctly philosophical, and philosophy is distinct, even if continuous with science. The no miracle argument is usually taken to explain scientific success by postulating the truth of theories, but both this explanandum and this explanans are found to be problematic. Instead, I take as explanandum the success of scientists in finding theories with unexpected predictions, and as explanans (so as conclusion of the argument) the actual finding of partially true theories, through reliable scientific method, thanks to the simplicity, uniformity and rationality of nature. Moreover, I claim that these explanations are not available to the various forms of idealism. Hence, while scientific and metaphysical realism are logically independent of each other, one can hardly support scientific realism by the " no miracle " without also accepting metaphysical realism.

Predictivists use the no miracle argument to argue that ‘‘novel’’ predictions are decisive evidence for theories, while mere accommodation of ‘‘old’’ data cannot confirm to a significant degree. But deductivists claim that since confirmation is a logical theory-data relationship, predicted data cannot confirm more than merely deduced data, and cite historical cases in which known data confirmed theories quite strongly. On the other hand, the advantage of prediction over accommodation is needed by scientific realists to resist Laudan’s criticisms of the no miracle argument. So, if the deductivists are right, the most powerful argument for realism collapses. There seems to be an inescapable contradiction between these prima facie plausible arguments of predictivists and deductivists; but this puzzle can be solved by understanding what exactly counts as novelty, if novel predictions must support the no miracle argument, i.e., if they must be explainable only by the truth of theories. Taking my cues from the use-novelty tradition, I argue that (1) the predicted data must not be used essentially in building the theory or choosing the auxiliary assumptions. This is possible if the theory and its auxiliary assumptions are plausible independently of the predicted data, and I analyze the consequences of this requirement in terms of best explanation of diverse bodies of data. Moreover, the predicted data must be (2) a priori improbable, and (3) heterogeneous to the essentially used data. My proposed notion of novelty, therefore, is not historical, but functional. Hence, deductivists are right that confirmation is independent of time and of historical contingencies such as if the theorist knew a datum, used it, or intended to accommodate it. Predictivists, however, are right that not all consequences confirm equally, and confirmation is not purely a logical theory-data relation, as it crucially involves background epistemic conditions and the notion of best explanation. Conditions (1)–(3) make the difference between prediction and accommodation, and account for the confirming power of theoretical virtues such as non ad-hocness, non-fudging, non-overfitting, independence and consilience. I thus show that functional novelty (a) avoids the deductivist objections to predictivism, (b) is a gradual notion, in accordance with the common intuition that confirmation comes in degrees, and (c) supports the no miracle argument, so vindicating scientific realism.

Nel primo capitolo preparo il campo alla discussione cercando di chiarire che cosa s'intenda per realismo in generale, che cosa distingua e che cosa accomuni i vari tipi di realismo di cui si parla in filosofia, e in particolare come si caratterizzi il realismo metafisico. Questo dovrebbe limitare il rischio di confusioni preliminari, derivante dal fatto che oggi si discute molto di realismo e di realismi, ma con terminologie differenti e spesso imprecise.
    Nel secondo capitolo esamino e critico le motivazioni, spesso implicite o appena sommariamente accennate, di chi ritiene che la confutazione kantiana del realismo metafisico sia di portata sufficientemente generale da mantenere la propria validità anche oggi che le specifiche argomentazioni di Kant sono generalmente rifiutate.
    Nel terzo capitolo difendo invece il realismo metafisico da un argomento ben più articolato (anche se non sempre esposto con la massima chiarezza e sistematicità): l'argomento delle descrizioni equivalenti di Goodman e di Putnam.
  La posizione che  difendo mantiene alcuni aspetti anche non secondari del criticismo kantiano. Tra l'altro, mi baso su di un importante presupposto che ha chiari precedenti nella Critica della ragion pura: la possibilità di distinguere, almeno in via locale o contestuale, tra forma e contenuto della conoscenza, o in termini più moderni tra fatti e convenzioni. Poiché questa possibilità è stata recentemente contestata da filosofi come Quine, Davidson, Rorty, e gli stessi Goodman e Putnam, nel quarto capitolo cerco di mostrare che i loro argomenti non sono realmente incompatibili col tipo di distinzione che a me interessa.

Apparently, neither the appeal to approximate truth nor to novel predictions can rescue the “no miracle” argument from Laudan’s criticisms. For Lyons (2002) there are historical counterexamples even to the weakened “deployment” realism: novel predictions supposedly derived from false claims. But if so, those successes would seem unexplainable, even by Lyons’ “modest surrealism”. In fact, I argue, some of those predictions were enough probable (either a priori or in the light of available evidence) to be guessed, and others were not essentially derived from false claims. Hence, a claim can be assumed to be true if it is actually crucial in deriving a prediction which is novel in a suitably specified sense.

The 2010 issue of Hermeneutica contains the proceedings of the conference Cos’è l’Illuminsmo? held in Urbino in 2009, with an introduction by Piergiorgio Grassi. The fourteen articles (plus the translations of an essay by D‟Alembert and one by Eberhard) explore both the many faces of Enlightenment and the manifold approaches to this phenomenon in past and present critical literature. Reflection inspired by this volume shows both the deep persisting relevance of Enlightenment to our present concerns, and its wide conceptual scope: the different names it is given in different languages may refer to (1) a historical movement, (2) its ideas, (3) an event or course of events, (4) the positive aspects of (1), (2), and (3), or (5) a meta-historical category. Moreover, these are typically cluster concepts, characterized by family resemblances, and the question “What is Enlightenment?” may be understood as (a) analytic or conceptual, or (b) empirical, or (c) evaluative. The relationships between Enlightenment and Christianity emerge as a paramount question, deeply involving our conceptions of the two relata. In particular, asking if Enlightenment is compatible with Christianity is of a piece with asking if it is self-consistent, or harbours the premises of its own overturning, as argued by Horkheimer and Adorno.

Evandro Agazzi’s volume Scientific Objectivity and its Contexts is here introduced. First, the genesis and the content of the book are outlined. Secondly, an overview of Agazzi’s philosophy of science is provided. Its main roots are epistemological realism in the Aristotelian/scholastic tradition, and contemporary science-oriented epistemology, especially in Logical Empiricism. As a result, Agazzi’s thought is nicely balanced between empiricism and rationalism, it avoids gnoseologistic dualism by stressing the intentionality of knowledge, and it insists on the operational and referential character of science. Finally, an account is given of Agazzi’s view of the origin and nature of scientific objects, which allows to understand how his sophisticated and “perspectival” realism differs both from naïve realism and constructivism.