In On Lutz on Laudan and demarcation, Paul Newall notes that:
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 the discussion with Lutz which ensued, I began to indicate reasons why necessary and sufficient conditions do indeed have to be taken into account if we are to demarcate – which is to say most soundly determine – both what is science/scientific and what is not science/scientific. This entry originally appeared in the discussion which occurred in the comments responsive to the Newall posting. The statements blockquoted below are remarks made by Sebastian Lutz in this comment. See the rest of that discussion for a more detailed context of what follows here.
1. I am really, really not assuming that the sufficient condition is a conjunction of necessary ones. Thus it is irrelevant that if I did assume that, I would be in agreement with Laudan.
6. I am really, really not assuming that the sufficient condition has to be an if-and-only-if condition (that is, also necessary).
It is true that Laudan’s conjunction could be assumed to be true with the argument against Laudan still succeeding. It has not been said that Lutz made that assumption, but it could be said that: a) logical demands require Lutz to use a combination of necessary and sufficient conditions, and b) if there are any differences between Laudan’s necessary-sufficient combination and a necessary-sufficient combination that Lutz might use to effect a valid argument, those differences are not to the detriment of Laudan’s argument.
2. I have as much in mind for sufficient conditions as I have for necessary conditions: Only their formal relations to ‘being scientific’. If something is a science only if it fulfills condition A, A is a necessary condition, and if something is a science if it fulfills condition B, B is a sufficient condition.
3. A sufficient condition that is not also necessary cannot determine that something is not a science. It can only determine that something is a science. This is, by the way, what Laudan says as well (pp. 118f).
Likewise, Laudan notes that a satisfied necessary condition which is not also sufficient is not enough to establish that something is a science. Hence, we have Laudan putting forth the combination of sufficient and necessary to satisfy both the determination of what is and what is not science/scientific. It is a prima facie legitimate hypothesis if only because, as is well known and as was discussed, the notion of a necessary-sufficient combination or conjunction is utterly conventional, and Lutz seems to have admitted that it is only with some sort of combination of sufficient and necessary conditions that it would be at all possible to determine both what is and what is not science/scientific.
5. Having a criterion with one sufficient and one different necessary condition does not mean that demarcation is subjective. For all those things that don’t fulfill the necessary condition, and for all those things that fulfill the sufficient condition, the criterion delivers a clear verdict, just as a necessary and sufficient condition would. For the unclear cases, there is, of course, no clear verdict.
For now, let us concentrate on the matter of “a criterion with one sufficient and one different necessary condition”; the notion of non-subjective demarcation (what I would more likely call judgment) can be addressed at some other time.
Given one particular context, something which satisfies the contextually sufficient condition must also satisfy the contextually relevant “different necessary condition”; otherwise, it is nonsensical to assert that the supposedly necessary condition is genuinely (or necessarily) necessary. In the case of a context of competing claims, it makes no sense to say that one claim for scientific status need not meet some necessary condition to qualify as scientific whereas an allegedly competing claim must meet that particular necessary condition.
Such an inconsistency (in this case an apparently inconsistent application of a supposed standard) is a telltale sign for possible logical invalidity. To assert that one claim to scientific status need not meet some necessary condition to qualify as scientific whereas a purportedly competing claim must meet that particular necessary condition is nonsensical precisely because it is illogical so long as the inconsistency remains apparent (or, in other words, so long as it is not explained, justified, or validly supported).
This is not to argue that there must only be one standard for all claims to scientific status (even though there might well be such an argument – which happens not to be essential for what is being argued here), but proponents for one claim to scientific status cannot logically relegate to unscientific status a competing claim by virtue of a standard which those proponents do not apply to their own preferred claim or which the proponents’ claim does not itself satisfy. After all, if neither of the competing claims satisfies the necessary condition, then it is either the case that neither claim is scientific, or it is the case that the condition is not actually necessary.
Accordingly, it appears that although Lutz did not assume “that the sufficient condition is a conjunction of necessary ones”, he does in fact rely on a necessary-sufficient combination — if his is a logical argument about incompatible competing claims.
Is reliance upon such a necessary-sufficient combination any different from Laudan relying upon “conditions which are both necessary and sufficient”? Laudan could just as well be said to be relying upon “both necessary and sufficient conditions”. Would this modification to Laudan’s expression change the meaning he intended?
Apparently not, because Laudan makes it clear that he is talking about individually necessary conditions which produce a sufficient condition when taken jointly, which is to say that Laudan is talking about the combination of necessary and sufficient conditions.
Still, there is that “jointly” matter. There seems to have been agreement in the referenced previous discussion that were a sufficient condition to be defined successfully in terms of all possible necessary conditions to be fulfilled, this would be the most precise sort of demarcation criterion. However, a weaker sort of demarcation could be one in which there is a sufficient condition in addition to – rather than jointly produced by – some necessary conditions. This really amounts to a different sort of “jointly”, but it is still correct to say that even under these circumstances it is the necessary and sufficient conditions taken in combination (or jointly) that provide the best available criterion for being able to designate or determine both what is scientific and what is not. This is to say once again that a necessary-sufficient combination (if attainable) will always produce a clearer distinction than will a sufficient condition alone or necessary conditions alone. (* See below for discussion about the possibility regarding demarcation in terms of unrelated theories.)
This is to say that, in the context of competing theories, Laudan’s combination claim has NOT been shown to be false. This leaves opponents in this context with only one available course of attack against Laudan, and that attack can either be thought of as being conducted in terms of the nature of objectivity, or it can be thought of as an argument over the definition of demarcation. And what I find most amusing about these repeated attacks on Laudan, at least thus far (and I am not here – or yet – including the Lutz argument as one of these attacks), is not that they amount to nothing more than a temper tantrum induced by highly demanding standards; what I find most amusing is not that these attacks do not themselves have anything to do with the activities within science (with what is effectively a unanimity, those involved in the sciences do not think in terms of or concern themselves with criteria for scientific/unscientific demarcation when they are designing and conducting those activities which they presume are scientific). No, what I would probably find most amusing about the attacks against Laudan is that they suggest a mind-set against philosophy – but that actually strikes be as being sad rather than amusing.
Laudan makes it quite clear that the demarcation matter is a non-issue since there are other means by which to deal with the problems that arise as a result of our inability to thus far devise an actually invariant demarcation criterion. The demarcation problem is essentially a run-of-the-mill epistemological problem, and further attempts to alleviate this problem might end up being helpful, but philosophy is concerned with wisdom that goes beyond mere knowledge, and the seeking of wisdom in no way depends upon – and it never demands – the weakening of epistemic demands or stringency.
So, science and philosophy are not lessened by Laudan’s “ideal”. Why, then, are there these repeated attacks on Laudan? For nearly thirty years, Laudan’s opponents, from Ruse through Pennock, et al., have explained the reason quite explicitly. Laudan’s argument is not readily applicable in certain legal/political contexts; Laudan’s argument does not serve certain legal/political interests.
* The earlier discussion concerned claims for scientific status by two incompatible competing theories, since the most common relevant context has been that which is set in terms of evolution and creationism/ID. There is also the possibility that the Lutz argument is supposed to pertain to different theories from different scientific fields. If one presumably scientific field were not to have (let us say) the same necessary condition(s) as a second presumably scientific field, and if a theory in the second field did not satisfy a necessary condition for that field, then that theory would not suffice for inclusion as part of that specific field. However, without additional argument, it is logically invalid to assert that a theory is unscientific on the basis of the failure of that theory to satisfy a necessary condition for one particular presumably scientific field. All that can be validly concluded in the absence of additional argument is only that the theory does not qualify as part of that particular scientific field; whether the unqualified theory is unscientific is another matter altogether. But, then, when we broaden the scope from the question of whether different theories suffice for their different intended fields to the question about whether the different theories are each alone or both together scientific or unscientific, we are back to having to apply to both theories the same standard in terms of a combination of necessary and sufficient conditions in order for us to be able to definitively and validly cover both the scientific and unscientific possibilities. It appears that Laudan’s point about the need for a necessary-sufficient combination remains intact across multiple conditions/contexts.