In his (draft) paper On an Allegedly Essential Feature of Demarcation Criteria of Science, Sebastian Lutz claims that demarcation does not require a criterion that is both a necessary and sufficient condition, as had been discussed in Laudan’s famous paper The demise of the demarcation problem. In a succinct counterargument, Lutz explains that we might have a sufficient condition allowing us to say that something is scientific and a distinct necessary condition allowing us to say that something else is not scientific; in short, that “Laudan’s demand that [the necessary and sufficient criteria] be one and the same is supererogatory”. Lutz goes on to propose that Laudan’s admission that most candidates for necessary criteria are implausible implies that some are not, and furthermore that Laudan’s comments suggest that, since we have an ordinary use sense of the demarcation between science and non-science, there must also be a plausible sufficient condition. Consequently, there may be a necessary condition and a separate sufficient condition for demarcating scientific theories.
I want to defend Laudan’s paper on several grounds. Firstly, it does not follow from our everyday use of demarcation that a philosophically rigorous sufficient criterion exists. Although we might, as Lutz posits, try to enumerate what we ordinarily call science, it is precisely this attempt that – along with the search for a necessary criterion – has failed. When Lutz argues that this ordinary criterion, although possibly very weak, can “already decide important cases”, the implication is that our ordinary use of the term science is enough to work with and build upon; but Laudan’s requirement of the enumeration is that it should be an “adequate explication” and “must exhibit epistemically significant differences”. We therefore need to show why our ordinary use provides a sufficient condition that will survive philosophical scrutiny, rather than starting from it.
Moreover, and on the other hand, if this ordinary use suggests that we have some kind of sufficient criterion, we should also be fair to Laudan’s invocation of it and note that our typical understanding of demarcation involves a criterion that is both necessary and sufficient. Lutz’s criticism is perhaps slightly uncharitable insofar as Laudan explicitly discusses a demarcation criterion that we assume will do the work of identifying both what is and what is not science. This captures our ordinary use of demarcation as a means of division rather than determining that something is x while separately determining that something else is not-y; that is, building upon our ordinary use. After all, this is what we see in examples like verification and falsification: the criterion is intended to demarcate for us on its own. This does not mean we should limit ourselves to seeking a single criterion but only that if we are going to begin with ordinary use then we should be consistent.
Secondly, Laudan does not claim that separate conditions can never be found because he does not need to. Indeed, this early part of Laudan’s paper is preparatory work: he reviews demarcation traditions, arguing that they have failed, but the discovery of a successful demarcation criterion would not imperil his main argument, which is not that the demarcation criteria so far identified have failed or that successful criteria can never be found but rather that demarcation is superfluous and beside the point.
I have explained this previously but demarcation’s demise is thus not due to the lack of a philosophically rigorous demarcation criterion but instead because it turns out that, for Laudan at least, demarcation is unhelpful and unnecessary. Much like the objection I made to the invoking of parsimony as a methodological rule, the merits or otherwise of an idea only become apparent after we have investigated it. Even those that seem straightforwardly to be hopeless may subsequently prove to have something to recommend them and we are extremely unlikely to discover this if we reject them ab initio. Note that this does not imply that all ideas will have such merits in due course, or that we should set aside time, money and other resources to conduct the investigation. If advocates of an idea wish to see it succeed, it is for them to pursue it and, when a case can be made, to present it in a stronger form.
For Laudan, this investigation would result in our being able to say whether or not the idea turned out to be well-founded and well-confirmed, but it could also be that philosophical examination provides additional criticisms or supporting arguments. Whatever the case, the idea or the theory it becomes is ultimately assessed based on how it survives this period of scrutiny. There is no need to pre-judge this process via demarcation because that removes or hinders the possibility that what seems like a bad idea might turn out to be useful or successful. Although this may sometimes be unpalatable if we want to give a final answer on ideas we dislike or have reasons to associate with ulterior motives, it is this openness that allowed science to develop and for some originally discredited theories to eventually prove their worth.
If it were the case that we have or could one day find a necessary and sufficient demarcation criterion, or if we could identify separate criteria as Lutz suggests, Laudan’s argument in The demise of the demarcation problem would not be threatened because, as he says in closing his paper, the status of claims is irrelevant; what really matters is their credentials. We should perhaps be careful not to put the cart before the horse: it is because an idea seems to have little to recommend it that we reject it, rather than seeking to reject or accept it from the outset. Although it is tempting to try to shortcut this process, science might turn out to be the poorer for it or, if this view had been applied in the past, not have developed at all.
Laudan, L. (1983) The demise of the demarcation problem, in R. S. Cohen & L. Laudan (Eds.), Physics, Philosophy, and Psychoanalysis (pp. 111–127). Dordrecht: Reidel.
Lutz, S. (2011) On an Allegedly Essential Feature of Demarcation Criteria of Science [Draft]. PhilSci Archive, http://philsci-archive.pitt.edu/8609/.