Against Verificationist Scientific Realism

At the end of Sect. 3.2, we raised the question: is verificationist scientific realism, scientific realism proper? The worry is this: Why should scientific realism incor­porate the claim of mind-independence (as elaborated by the possibility of diver­gence)? Why, that is, couldn’t someone who accepted the reality of unobservable entities without also rendering them mind-independent (in the above sense) be a scientific realist?

Crispin Wright has aimed to show that a verificationist version of scientific realism is indeed possible and desirable.

This further content of realism, Wright argues, is consistent with verificationism since verificationism need not entail “a thesis about the bounds of reality—the thought that, as it were, the totality of facts is conveniently (but mysteriously) trimmed to ensure that there is nothing there that outreaches human inquisitiveness” (ibid.).

Well, we have already seen Dummett claiming that “the world is, so to speak, formed from our exploration of it”, which might well imply that there is nothing beyond human inquisitiveness when it comes to how far the content of the world extends.

Indeed, Wright takes it to be the case that this verificationist version of scientific realism would require that the “faithfulness” of scientific representation—that is, its representing things as they are—is “in principle detectable”. And though he is right in stressing that, on this view, there is “nothing to foreclose on the thought that scientific theory deals in genuinely representational but irreducible contents, ren­dered true or false by a real external source” (Wright 1992, 159), this very claim only shows that verificationist scientific realism is not instrumentalism. The key point here is that Wright's verificationist scientific realism renders the world that science probes mind-dependent in precisely the sense we have been describing, viz., it forecloses the possibility of a divergence between what there is in the world and what is in principle known (detectable) to exist. It compromises objective existence even if it honours irreducible existence.

Note that an appeal to idealized (or in-principle) conditions for justification is a double-edged sword.

This tension is seen in Wright's favourite version of an epistemic account of truth, which relies on a strong account of justification. As is well known, Wright (1992) introduced the idea of superassertibility as a candidate for an epistemic notion of truth. Superassertibility is “assertibility which would be endurable under any possible improvement to one's state of information” (Wright 1992, 75). Only such a strong epistemic conception rules out the possibility that a warrant- edlyassertible statement will not be controverted by subsequent information and hence enshrines the stability of truth. Interestingly, Putnam (2001, 599) claimed that this was exactly the notion of truth that he had in mind in his verificationist years.

To see how strong this requirement of incontrovertibility of justification is, it is enough to state that the only way to ensure it is to claim that the available evidence entails the truth of a certain statement S; otherwise, it is always possible that there may be further or future evidence that will remove the warrant for the truth of S. This requirement then would end up implying either that scientific hypotheses are non-ampliative or that we are already in possession of a priori justifications of ampliative methods such as induction or inference to the best explanation. But we are far from this, if it is possible at all.

Superasseitibility is meant to close the gap between truth and incontrovertible warranted assertion.

Note that the very possibility of a verificationist scientific realism requires the a priori falsity of the underdetermination of theories by evidence, as Wright himself recognises (cf. Wright 1993, 287). It is obvious that the very logical possibility of two or more mutually incompatible theories being empirically equivalent entails (on the assumption that only one of them can be true) that truth doesn't necessarily lie within our cognitive capacities and practices. Moreover, if it is possible that there are incompatible but empirically equivalent alternatives even to an idealised theory of the world, it cannot be the case that an idealised theory of the world is necessarily true; hence the world might differ from the way it is described to be even by an ideal theory of the world.

Though scientific realism must find ways to block the argument from underdetermination of theories by evidence in order to ground the epistemic opti­mism associated with the third thesis (see the Introduction), it is questionable that it should be committed to the a priori falsity of the underdetermination thesis (cf. Psillos 2015). The reason for this is precisely that by doing so, the independence of the world is compromised—in particular, the objective existence condition. As Newton-Smith (Newton-Smith 1978, 88) has nicely put it, presented with the possibility of underdetermination, realists face a dilemma. Realists can go for an ‘Ignorance Response' or an ‘Arrogance Response'. On the first horn, realists choose to cling to a realist metaphysics of an independent world, but they put at stake their epistemic optimism. On the second horn, they secure epistemic optimism, but sacrifice the independence of the world by endorsing a view which denies that there are ‘inaccessible facts'.

So is the ‘ignorance response' a compelling move for realism? Realism, to be sure, makes scepticism possible. This is the price that realism has to pay for respecting the thesis that the world is mind-independent. In fact, a key motivation for the anti-realist denial of mind-independence has been blocking of scepticism. But though ignorance is possible, realism does not have to argue that it is actual! In fact, the standard argument for scientific realism, known as the Putnam-Boyd argument from the success of scientific theories to their truth, is precisely meant to ground the realist epistemic optimism on the contingent fact that the (approximate) truth of the scientific theories is the best explanation of their empirical successes (cf. Psillos 1999, 2009).

In characterising non-verificationist scientific realism, Wright says: “If it is as likely as not, for all anyone can have cause to believe, that prosecution of best method leads to false theories as to true ones, what is the basis for believing that we can even so much as get lucky?” (Wright 1993, 292). This kind of question, and the implicit point made by it, totally disregard the fact that scientific realists have argued precisely for the point that the best explanation of pursuing successfully best method is that the relevant scientific theories (that inform best method) are approximately true.

When it comes to the argument from underdetermination, the right realist answer is to argue that the tie between competing empirically equivalent rival theories is broken by looking for reasons—mostly explanatory reasons—to prefer one theory over the rest. In other words, look for the best explanation of the evidence. It is not the purpose of this paper to revisit the debates about the so-called ‘no miracles' argument or about the underdetermination thesis (see my Psillos 1999, 2015). It is enough for our purposes to note the following. A key motivation for verificationist scientific realism is that it renders scientific truth achievable. As Wright (Wright 1993, 298) put it, it wields an essential connection between “the harvest of best scientific method and truth”. What it adds to ordinary scientific realism is that “truth is essentially certifiable by best method: that for any true theory, sufficiently extensive researches must disclose an adequate, enduring case for taking it to be so”. Beware though: on the verificationist approach, it is not that the method harvests an independently given truth about the world; the truth is the deliverances of the method (suitable and persistently pursued). Hence, there is no possibility of divergence; hence, the mind-independence of the world (in the form of the objective existence condition) is compromised.

What, ultimately, is at stake in the scientific realism debate is whether there can be a robust sense of objectivity, that is a conception of the world as the arbiter of our changing and evolving theoretical conceptualisations of it. The kernel of the metaphysical thesis of scientific realism is that science is in the business of dis­covering what a world that is not of our making is like. This thesis implies that if the natural kinds posited by theories exist at all, they exist objectively, that is, inde­pendently of our ability to be in a position to know, verify, recognise etc. that they do; and hence that it is they, if anything, that make scientific theories true. Veri- ficationist scientific realism compromises this robust sense of objectivity.

Moreover, a moral that might be drawn from the debates around scientific realism is that the success of science—that realism is meant to explain—is hard won. It is neither trivial, nor in any way guaranteed. The heated debate over the pessimistic induction (see my Psillos 1999, Chap. 5) has driven the point home that if there is continuity in theory-change, this has been a considerable achievement, emerging from among a mixture of successes and failures of past scientific theories. A realist non-epistemic conception of truth and in particular the possibility of divergence do justice to this hard-won fact of empirical success and convergence. Given that there is no a priori guarantee that science converges to the truth, or that whatever scientists come to accept in the ideal limit of inquiry or under suitably ideal epistemic conditions will (have to) be true, the claim that science does get to the truth (based mostly on explanatory considerations of the sort we have already canvassed) is quite substantive and highly non-trivial. If, on the other hand, an epistemic conception of truth is adopted, hence if the possibility of divergence is denied, the explanation of the success of science becomes almost trivial: (long-run) success is guaranteed by a suitably chosen epistemic notion of truth, since—ultimately—science will reach a point in which it will make no sense to worry whether there is a possible gap between the way the world is described by scientific theories and the way the world is.

Wright has called ‘absolute realism' the view that is anti-instrumentalist, “but drops the epistemological optimism with which scientific realism is usually asso­ciated (Wright 1993, 292). This position, he adds, makes the connection between scientific theoretical truth and the deliverances of best method “inscrutable”. Absolute realism, in other words, is non-verificationist realism and the remedy that is suggested to bring back epistemic optimism is to adopt the view that all truth is detectable (and harvested, in principle, by the scientific method). I have already stressed that what Wright calls absolute realism is not the position associated with current scientific realism. But not because scientific realism adopts an epistemic notion of truth; but because it need not abandon epistemic optimism. All it needs to do (and defenders of realism have done) is to show that there is a hard-won contingent link between the harvest of scientific method and truth. Cognitive success might is a contingent matter; but it is no less success for this.

Let me grant, at least for the sake of the argument, that, as a matter of contingent fact, whatever is issued by an epistemically right theory of the world (that is a theory which meets some ideal epistemic constraints) is what really exists in the world. This is certainly a possibility. It is consistent with the possibility of divergence noted above, and does not compromise the mind-independence of the world in any sense. Nor does it commit scientific realism to an epistemic account of truth. The scientific realist can easily accommodate the envisaged possibility of convergence by taking the right side in the relevant Euthyphro contrast: is the world what it is because it is described as thus-and-so by an epistemically right (meaning: suitably justified, superassertible etc.) theory or is a theory epistemically right because the world is the way it is? At stake here is the order of dependence. Scientific realists should go for the second disjunct, while verificationist anti-realism goes, ultimately, for the first.

Hence, it is wrong to pose to realist the following dilemma: either the concept of truth should be such that cognitive success is guaranteed or else any cognitive success is a matter of pure luck. What the realism debate has taught us, to be sure, is that the success of the realist project requires some epistemic luck: if the world were not mappable, science would not succeed in mapping it. But the realist has a story to tell us to why and how cognitive success, though fortunate and not a priori guar­anteed, is not merely lucky or a matter of chance. The realist story (see Psillos 1999) will have to be phrased in terms of the reliability of scientific method and its defence. But there is good reason to think that this story is both sensible and credible.