by Lee McIntyre
To explain this problem more precisely, one must understand the power of the biconditional relationship in logic. To say “A if and only if B” is equivalent to saying “B if and only if A.” So to say that “a theory is scientific if and only if it is falsifiable” is equivalent to saying “a theory is falsifiable if and only if it is scientific.” But this means that “being falsifiable” and “being scientific” are logically equivalent. Once you put “if and only if” into the sentence, “science” and “falsifiability” have identical truth conditions.7 (See table 4.1.) A necessity standard is too strong and a sufficiency standard is too weak. But by adding them together, instead of creating a criterion that is “just right,” the problems seem to multiply. Now the standard cannot satisfy our intuition that evolutionary biology is scientific but astrology is not.8
Table 4.1
Criterion
Type
Problem
If science, then falsifiable
(If not falsifiable, then not science)
necessity
Is evolutionary biology not science?
If falsifiable, then science
(If not science, then not falsifiable)
sufficiency
Is astrology science?
Science iff falsifiable
(Falsifiable if and only if science)
equivalence
Is astrology science, but evolutionary biology is not?
This is why I renounce the “necessary and sufficient conditions” approach when discussing the scientific attitude as a means of understanding what is special about science. For if one were to try to put the scientific attitude forward as a set of necessary and sufficient conditions, I think that the same type of problems would arise as we just saw with Popper. In particular, defending the distinctiveness of science does not seem to require a sufficiency condition. It is enough to embrace the necessity condition and say “For an area of inquiry to be a science, it must have the scientific attitude,” which is logically equivalent to saying “If a theory does not have the scientific attitude, then it is not science.” (See table 4.2.)
Table 4.2
Criterion
Type
If science, then scientific attitude
(If not scientific attitude, then not science)
necessity
If scientific attitude, then science
(If not science, then not scientific attitude)
sufficiency
Science iff scientific attitude
(Scientific attitude iff science)
equivalence
Of course, I realize that by choosing not to specify a sufficiency condition we will have only identified what is not science, not what is. But perhaps that is all one needs. We won’t be able to say definitively whether something like string theory is scientific. Or even the rest of physics. But is that really so problematic? Although we may have started out wondering whether there was some way to show why science was scientific, perhaps that is the wrong question. Maybe the goal of saying what is distinctive about science is instead to show why some areas of inquiry are not scientific. We may define science by its absence. This allows us to protect it from imposters. When we discover the essential properties of science, we can use them to say that if something does not have them then it cannot be science. But this does not mean that everything that has the scientific attitude is a science. It is a necessity standard, not a sufficiency one.
Ever since Laudan, those who have tried to provide a criterion of demarcation have run aground on the issue of trying to provide both necessary and sufficient conditions. Not only can it probably not be done (witness the history of the problem both before and since Laudan’s “obituary”), but there are serious costs in trying to follow Laudan’s mandate. Instead, I propose that we can get the job done by looking only for the necessary conditions for science, one of which is the scientific attitude. Once we have said that if we have science, then we must have the scientific attitude—and have thus found a necessary condition for science—we have also said that if we do not have the scientific attitude, then we do not have science—which turns out to be the sufficiency condition for nonscience. When we say “If an area of inquiry does not have the scientific attitude, then it is not science,” the logical form is “If not SA, then not science,” which by contraposition is equivalent to “If science, then SA.” (See table 4.3.)
Table 4.3
Case 1
If science, then SA
Necessity standard for science
is equivalent to
is equivalent to
If not SA, then not science
Sufficiency standard for nonscience
Case 2
If SA, then science
Sufficiency standard for science
is equivalent to
is equivalent to
If not science, then not SA
Necessity standard for nonscience
One must be careful here. It is tempting at this point to want to complete the second half of the job and try to mold the scientific attitude into a sufficiency standard for science, so that we can solve the traditional problem of demarcation. But we must avoid this. While the equivalences outlined in table 4.3 are correct, the tasks in case 1 and case 2 are entirely separate. I contend that case 1 is all we need to identify what is special about science. (And of course, as logicians know, case 1 does not in any way imply case 2.) There would be a high cost for engaging case 2 and trying to put forth the scientific attitude as a sufficiency standard for science. Not only would this pull us back into Laudan’s quagmire of specifying the necessary and sufficient conditions for science (which would mean defending the assertion that the scientific attitude is logically equivalent to science), it would also entangle us in the vexed question of how many different ways something can fail to be scientific.
Remember that the sufficiency standard for science is the necessity standard for nonscience. But why should we care at all about the necessity standard for nonscience? There are myriad ways for an inquiry to fail to be scientific. Art and literature are not scientific, primarily because they are not trying to be. Astrology and creationism are a different story; here their subjects of inquiry do seem to be empirical, so we should care very much about how they fail to be scientific. All of these fields may lack the scientific attitude, but are we really ready to say that this is a necessary condition for nonscience? All one has to say is that if these fields do not embrace the scientific attitude—which none of them do, for different reasons—they are not sciences. Our current project requires only that lack of the scientific attitude is sufficient for being nonscience, not that there is only one way of being nonscientific. This may be more of a live question when we are interested in some of those fields in social science that want to be considered as sciences but are not yet up to the standard. Yet here again our original insight pays off: there may be many ways of being nonscience, but if one wants to become scientific there is only one way: it is necessary to embrace the scientific attitude. Thus I submit that one can say something interesting about what is special about science without solving the traditional problem of demarcation.
It is important here to be sensitive to the issue of condescension in judging what is and is not science. In contrast to some who have wrestled with the problem of what is special about science, I maintain that the scientific attitude need not force us to say that all of nonscience is an inferior human pursuit. (Indeed, it had better not be, because I would count philosophy in that camp!) Literature, art, music—all are nonscience. They do not have the scientific attitude, but that is completely expected. The problem instead comes from some nonscientific disciplines that are trying to present themselves as scientific. With astrology, creationism, faith healing, dowsing, and the like, we encounter disciplines that are making empirical claims but refuse to abide by the practices of good science, one of which is that they have the scientific attitude toward evidence. They are, in short, only pretending
to be scientific. Here it seems more appropriate to scorn those fields that are masquerading as sciences or claim to have some special access to empirical knowledge outside the channels of scientific practice. When one embarks upon the pursuit of empirical knowledge, it is a serious problem to eschew the values of science.
Can One Still Try to Demarcate Science From Pseudoscience?
Some may object that all I have really done thus far is push the problem of demarcation back a step. For some have approached the problem of demarcation not as a dividing line between science and nonscience, but between science and pseudoscience. The virtue of this approach is that it goes directly to where the problem lies: those fields that are making empirical claims but only pretending to be sciences. So should I try to use the scientific attitude as a criterion for demarcating between science and pseudoscience and just cut off the whole question of the other nonsciences like literature and philosophy that are not making empirical claims? Perhaps some might wish to do this, by adding criteria to the scientific attitude such as “is making empirical claims” and “is not pretending to be scientific”9 in search of an attempt to solve the problem of demarcation between science and pseudoscience. But I think there are deep problems in proceeding this way.
As noted in chapter 1, if we look at the literature on demarcation we find a troubling equivocation on the issue of whether science is supposed to be juxtaposed with nonscience or with pseudoscience. The Logical Positivists seemed to be concerned with distinguishing science from metaphysics (which one might think of as a type of nonscience that the positivists wished to dismiss with some prejudice). Karl Popper, in his Logic of Scientific Discovery, seems to be concerned with nonscience, but by the time he writes Conjectures and Refutations, he characterizes his enemy as “pseudoscience.” Even Laudan equivocates on whether it is nonscience or pseudoscience that is the nemesis of science.10
Yet this makes a difference. Given the logical issues we have just explored surrounding necessary and sufficient conditions, it is imperative to understand science by looking at all of the things that are not science, not just those that are pretending to be. Yes, it is tempting to reserve special opprobrium for the frauds and denialists, the poseurs and the charlatans, and in chapters 7 and 8 I will do precisely this. But in considering the question of demarcation vis à vis the provision of necessary and sufficient conditions, the logic is unyielding: every inquiry must either be science or nonscience. The necessity conditions for science are the sufficiency conditions for nonscience, not pseudoscience. While it is true that a good deal of contemporary work on the demarcation problem since Laudan interprets the demarcation debate as between science and pseudoscience, this is problematic. As I argued previously (in chapter 1), both Hansson’s and Boudry’s accounts founder on precisely this failure to distinguish between nonscience and pseudoscience.11
I believe the correct distinction is this: the problem of demarcation properly concerns science versus nonscience. The category of nonscience includes—among other things—domains that are unscientific (such as math, philosophy, logic, literature, and art) that do not wish to make empirical claims, and those that are pseudoscientific (such as astrology, intelligent design, faith healing, and ESP) that do wish to play in the empirical sandbox, even while they flout the standards of good evidence.12 Other fields, such as ethics and religion, are harder to categorize. Some scholars have held, for instance, that ethics is completely normative and thus should join the rest of philosophy in the “unscientific” category. Others contend that parts of ethics make pretensions to science—or even fulfill this desideratum—so its status as either science or pseudoscience depends on your point of view. Religion, insofar as it makes empirical claims on spiritual grounds, is easier to classify as pseudoscience, yet some contend—as Galileo did—that religion has no business in the empirical realm and so should be considered unscientific.13 Much as I identify with the desire to heap scorn on the deficiencies and dishonesty of pseudoscience, a more straightforward approach is to recognize that the scientific attitude does its job both for identifying why literature and art are not science (because they care about empirical evidence only to the extent that it underlies their basis for creative expression and not to prove or disprove scientific theories) and why astrology and creationism are not science (because they are only pretending to care about empirical evidence and are unwilling to revise their theories). It is two birds with one stone. We do not need more than this.
Should I have confined myself in this book to discussing only those fields that sought to make empirical claims and thus attempt a demarcation between science and pseudoscience? For the reasons already given, I have chosen the logically cleaner route. Why search for additional necessary criteria for science (at best) or sufficient criteria (at worst), when the scientific attitude already does the job of specifying what is special about science? Thus I do not think that I have merely pushed the problem of demarcation back a step. I never promised to demarcate between unscientific fields and pseudoscientific ones—between literature and creationism, say.14 It also seems a virtue of my approach that it allows a path for those nonscientific fields that are not pseudoscientific (yet perhaps even for these) to emulate the high standards of science.
This book not only defends science, it proselytizes for it. Thus, I seek to make room for those fields that may genuinely care about empirical evidence—like the social sciences—but may not have always been true to the scientific attitude, perhaps because they did not realize that this was the essential ingredient in becoming more rigorous. I also believe that embracing the scientific attitude may help to turn a once pseudoscientific field into a successful science, much as occurred in the history of medicine.15 To understand what is special about science one need not scold or condemn all that is nonscience. To say that qualitative sociology is nonscience is scorn enough; one does not have to lump it in with witchcraft. In reading Popper, one sees him take delight in dismissing the work of Adler, Freud, and Marx in the same breath as astrology. Yet perhaps Freud, for instance, just did not recognize the empirical standards that his inquiry was required to live up to and could perhaps emulate as he sought to become more scientific. If philosophers of science took more care in identifying the necessary conditions for science, could even Freudian psychology be rehabilitated into embracing the scientific attitude?16 Insofar as possible, I want to remain nonjudgmental about those fields that are nonscientific. The scientific attitude is a standard on the empirical battlefield that may be picked up by any discipline that wants to become more rigorous.
Of course, many questions remain. In chapter 5, I will deal with the question of who judges whether someone has the scientific attitude and against what standard. In chapter 8, I will discuss how to handle those who fail to achieve the scientific attitude, either deliberately or in the false belief that they already embrace it, even though it is evident to the scientific community that something is amiss. At present, however, it is important to deal with two remaining questions that will be on the mind of the demarcationist. First, would it be so terrible if I tried to turn the scientific attitude into a sufficiency standard in addition to a necessary one? And second, even if I am not willing to do this, couldn’t I still make a claim to have solved the problem of demarcation by redefining it?
Should Everyday Inquiry Count as Science?
We have already dealt with the fallout from refusing to use the scientific attitude to demarcate between science and pseudoscience, and the consequence that this lumps some disciplines like literature and philosophy in with astrology and creationism on the nonscientific side of the ledger. Earlier we faced the fact that a necessity standard alone will not allow us to say that any given discipline is science, only that some are not. When we refuse to specify a sufficiency standard for science, this is the price we pay: does reliance on the scientific attitude solely as a necessary standard force us to rule out too much? Call this the problem of exclusivity. But what about the other half of the pr
oblem? What if we were suddenly tempted to try to offer the scientific attitude as a sufficiency standard—in addition to a necessary one—and thus take a shot at solving the problem of demarcation? This would create problems of its own. Remember Popper: by offering falsifiability as both a necessary and sufficient criterion for science, he did not dissolve his problems but doubled them. The same thing would happen here. If we said that having the scientific attitude was sufficient for being scientific, we would run the risk of permitting all sorts of random empirical inquiries into the pantheon of science. Would demarcationists be happy to consider plumbing and TV repair as sciences? Call this the problem of inclusivity.
One salient example—which illuminates those mundane occasions when we may be thinking scientifically but arguably not “doing science”–is looking for one’s keys.17 Suppose that my brother has lost his keys. He knows that he last had them in the house, so he does not bother searching the yard or going out to his car. He had them when he unlocked the front door, so they must be inside the house somewhere. First, he searches all the most likely places—his pants pockets, in the hall drawer, next to the television set—and then engages in an ever-widening search whereby he retraces his steps since he came in the front door. He starts in the hallway, goes to the bathroom where he washed his hands, and then into the kitchen where he made a sandwich before walking to the sofa to sit down and eat. There he finds his keys between the sofa cushions. It could be argued that in this activity he was employing the scientific attitude. He cared about empirical evidence and was attempting to learn from observation (as he looked in various locations and ruled them out one by one). He was also continually changing his hypothesis (“the key might be in X location”) as he went. But is this science?
