1. Introduction
It was Karl Popper who introduced the term “demarcation” for the identification of claims that should not be classified as scientific. His first published reference to this task can be found in a two-page letter to the editor of Erkenntnis in 1933.Footnote 1 There he referred to the problem of how to perform such a task as the “Abgrenzungsproblem” (later translated as “demarcation problem”) and said that this problem “can be defined as the request for a criterion to differentiate between ‘empirical-scientific’ and ‘metaphysical’ assertions (sentences, systems of sentences)” (Popper Reference Popper1933, 426).Footnote 2 He also briefly introduced the criterion of falsifiability as a solution to this problem. Pseudoscience was not mentioned in this text. He treated the topic much more extensively in Logik der Forschung (Popper Reference Popper1935), where he described the demarcation problem as that of finding a criterion to “distinguish between the empirical sciences on the one hand, and mathematics and logic as well as ‘metaphysical’ systems on the other” (Popper Reference Popper1935, 7; Popper Reference Popper1959, 34). Again, pseudoscience was not mentioned.Footnote 3
Immanuel Kant seems to have been a source of inspiration for Popper’s choice of terminology. In his 1933 letter to the editor, he referred to the demarcation problem as Kant’s question about the distinction between empirical-scientific and metaphysical sentences. Kant did indeed use the phrases “borders of natural science” and “borders of science” in Critique of Pure Reason (Kant [1787] Reference Kant1911, 488 and 540).Footnote 4 It should also be mentioned that members of the Vienna Circle occasionally used a similar terminology for various purposes. Moritz Schlick wrote in 1926 on how to demarcate (“abgrenzen”) metaphysics from empirical science (Schlick Reference Schlick1926, 153). In the 1929 “manifesto” for the Circle, Rudolf Carnap, Hans Hahn and Otto Neurath claimed to have “drawn the border of the contents of legitimate science” (Hahn et al. [Reference Hahn, Neurath, Carnap, Stöltzner and Übel1929] Reference Hahn, Neurath, Carnap, Stöltzner and Übel2006, 15).Footnote 5 In 1931 Otto Neurath wrote about how to demarcate (“abgrenzen”) the natural sciences from the humanities (Neurath Reference Neurath1931, 406). Against this background, Popper’s choice of terminology does not seem to have been particularly surprising or innovative.
When returning to the subject in a lecture in 1957, Popper gave a distinctly different account of the purpose of his criterion of demarcation. He now said that he had already grappled with the demarcation problem in the autumn of 1919. “I wished to distinguish between science and pseudo-science” he said, “knowing very well that science often errs, and that pseudo-science may happen to stumble upon the truth” (Popper Reference Popper and Alec Mace1957, 155). Deliberations on the weaknesses of Marxist historiography and Freudian and Adlerian psychology had contributed to his conclusion that falsifiability was the appropriate “criterion of the scientific status of a theory” (Popper Reference Popper and Alec Mace1957, 160). It solved “the problem of drawing a line (as well as this can be done) between the statements, or systems of statements, of the empirical sciences, and all other statements—whether they are of a religious or of a metaphysical character, or simply pseudo-scientific” (Popper Reference Popper and Alec Mace1957, 162). In an essay published in 1962, entitled “The demarcation between science and metaphysics,” he specified the purpose of his criterion more precisely as “drawing a line of demarcation between those statements and systems of statements which could be properly described as belonging to empirical science, and others which might, perhaps, be described as ‘pseudo-scientific’ or (in certain contexts) as ‘metaphysical’, or which belonged, perhaps, to pure logic or to pure mathematics” (Popper Reference Popper1962, 255).
In spite of Popper’s account in 1957 of his motives a quarter century earlier, his texts show a clear change in the proposed application of the falsifiability criterion. He introduced it as a criterion for distinguishing between (empirical) science and metaphysics, but later he presented it primarily as a criterion for drawing a line between science and pseudoscience. These are of course two different tasks.Footnote 6 Although Popper’s proposal for how to distinguish between science and pseudoscience has lost much of its popularity, his choice of terminology is still highly influential in the philosophical discussion on what assertions should be classified as pseudoscientific. This discussion has been almost exclusively couched in terms of the “demarcation” of science and/or pseudoscience. This is somewhat surprising since philosophers otherwise tend to prefer complete definitions of the central concepts of a discussion. Such definitions are usually expressed in terms of some set of conditions that are each necessary and jointly sufficient for inclusion in the defined category. In contrast, to demarcate between two categories A and B we only need to be able to distinguish them from each other. Therefore, a demarcation between two categories can be based on information that is insufficient for defining either of them.
Metaphors other than “demarcation” have also been used in the literature on pseudoscience. This includes terms such as “hinterland(s)” (Boudry Reference Boudry2011), “borderlands” (Shermer Reference Shermer2001; Moore Reference Moore2016), and “backwater” (Harmon Reference Harmon2012, 56), which refer to a spatial analogy but do not indicate a sharply defined border. It also includes terms that elaborate on the meaning of the prefix “pseudo-,” such as “mimicry” (Blancke et al. Reference Blancke, Boudry and Pigliucci2017, 78), “imitation” (Boudry Reference Boudry2022, 83), “counterfeit” (Jarosz Reference Jarosz2014), “forgery” (Balogh Reference Balogh2022, 871), “fake” (Schiele Reference Schiele2020, 3) and “sham” (Andre Reference Andre2023, 172). However, these metaphors are all fairly rare, and none of them can be described as established terminology. They will therefore not be discussed here.
The purpose of this article is to investigate whether the focus on demarcations rather than definitions has made any difference in the philosophical discussion on pseudoscience, and to clarify whether anything can be gained by instead conducting this discussion in terms of definitions.Footnote 7 Although there is a large philosophical literature on metaphors, only a small part of it refers to the use of metaphors in philosophy itself.Footnote 8 John Locke warned against adorning philosophical discourses with “lively metaphorical representations” that make it more difficult to “perceive what are the true ideas, upon which the inference depends” (Locke [1690] Reference Locke and Woolhouse1997, 597, Bk 4, Ch. 17). Kant pointed out that philosophical language is full of words that induce us to think in terms of another concept than the one that is directly applicable (Kant [1790] Reference Kant1913, 352–53). Paul de Man (Reference de Man1978) called attention to the ubiquity of metaphors and figurative language in philosophy. George Lakoff and Mark Johnson argued that metaphors are constitutive in philosophy and cannot be disposed of without loss of philosophical content (Lakoff and Johnson Reference Lakoff and Johnson1999, 339). The degree to which the use of metaphors influences philosophical thinking is difficult to assess, but empirical evidence from other areas strongly suggests that metaphors can have a considerable influence on human thought and behavior (Landau et al. Reference Landau, Chen-bo and Swanson2018; Jimenez et al. Reference Jimenez, Arndt and Landau2021; Hauser and Fleming Reference Hauser and Fleming2021; Condello Reference Condello, Wagner and Marusek2023). It is therefore reasonable to assume that widespread use of a metaphor in philosophical discussions can contribute to shaping or perpetuating the philosophical thought patterns that it supports.
Section 2 summarizes the usage of the terms “demarcate” and “demarcation” in other contexts than that of the distinction between science and pseudoscience. This section is based on an extensive search for texts that use these terms. A primary search in Philosopher’s Index, Web of Science, and Google Scholar was followed by a secondary search for texts cited in the sources that were found in the first round. The aims were to identify as many different uses of the terms as possible and to find texts that contribute to clarifying what is meant by demarcation in the various contexts. Based on these findings, Section 3 explores the potential consequences of performing a conceptual analysis in terms of demarcations rather than definitions. In Section 4, this is applied to demarcations and definitions of science and pseudoscience. Section 5 concludes.
2. The uses of “demarcation”
According to the Oxford English Dictionary, the term “demarcation” was first used about the meridian chosen by Pope Alexander VI in 1493 as a border between the parts of America to be subjugated by Spain and Portugal respectively (Anon 2024; Bown Reference Bown2011). Since then, the term has been used to denote geopolitical borders and the acts of setting them. Aside from the science–pseudoscience context, this still seems to be its most common usage (Latham Reference Latham1848; Brigham Reference Brigham1919; Howay et al. Reference Howay1921; Platt Reference Platt1924; Ryder Reference Ryder1925; McMahon Reference McMahon1935; Boggs Reference Boggs1937; Prescott and Triggs Reference Prescott and Triggs2008; Ben-Bassat and Ben-Artzi Reference Ben-Bassat and Ben-Artzi2015; Goettlich Reference Goettlich2022).
Since the early twentieth century, “demarcation” has increasingly referred more specifically to “the construction of boundary markers in the landscape” (Prescott and Triggs Reference Prescott and Triggs2008, 12), a process that typically follows the prior determination of the border in terms of coordinates or lines on a map. This is now how the term is used by international courts (Prescott and Triggs Reference Prescott and Triggs2008, 146; cf. Goettlich Reference Goettlich2022). The terminology seems to have been proposed in 1897 by the British officer and diplomat Henry McMahon. He had discovered that the distinction between determining a boundary and marking it physically “was not yet recognised, and that there were no two terms in common use by which to define them” (McMahon Reference McMahon1935, 4). He chose the term “delimitation” for “the definition on paper, either in words or on a map – of the limits of a country” and “demarcation” for physically marking the boundary on the ground (Rushworth Reference Rushworth1997, 1). The term “demarcation” was officially used in this sense in 1918 (International Boundary Commission 1918, 217 and 238; cf. Platt Reference Platt1924; Ryder Reference Ryder1925). McMahon’s term “delimitation” for determining a boundary on paper is also used today in international courts (Prescott and Triggs Reference Prescott and Triggs2008, 146). The words “definition” (Platt Reference Platt1924; Boggs Reference Boggs1937) and “delineation” (Ben-Bassat and Ben-Artzi Reference Ben-Bassat and Ben-Artzi2015; Smith Reference Smith and Tambassi2019, 145) have been used with the same meaning.
“Demarcation” can also denote other spatial limits than boundaries between countries. It has been used about the limits around areas that constitute informal settlements (“slums”; David and Busisiwe Reference David and Busisiwe2019), the limits between sedimentary strata (de Beaumont Reference de Beaumont1831), and the lower limit of the stratosphere (Hoinka Reference Hoinka1997). In medicine, the phrase “line of demarcation” has long been used for a sharp limit between a gangrene and adjacent living tissue (Bottomley Reference Bottomley1845; Garlock Reference Garlock1931; Latz et al. Reference Latz2022). In botany, the same phrase is used about a distinct line between sapwood and heartwood (Eades Reference Eades1937; Wang et al. Reference Wang, Zhang and Zhou2021).
Another common usage of the term refers to distinctions between different topics, for instance between ancient astronomy and ancient physical theory (Bowen Reference Bowen2007), physics and chemistry (Hiebert Reference Hiebert1996), logic and mathematics (Weingartner Reference Weingartner1982), philosophy of mind and cognitive science (Fumerton Reference Fumerton2007, 67), and philosophy and science (Díez Reference Díez2010). The term is also often used about distinctions between subject areas or types of discourses in a more general sense, for example between spiritual and secular issues (Hodge Reference Hodge1844, 545), moral and non-moral issues (Miller Reference Miller2020), federal and state legislation (Groner Reference Groner1928), the law of slander and that of libel (Carpenter Reference Carpenter1928), fictional and non-fictional historical narratives (Tyynelä and De Mey Reference Tyynelä and De Mey2012), and metaphorical and literal statements (Engstrøm Reference Engstrøm1996).
“Demarcation” can also refer to distinctions between different categories of human actions or behavior, such as those between civil(ized) and uncivil(ized) human behavior (Leerssen Reference Leerssen1995; Braunstein Reference Braunstein2018), felonies and misdemeanors (Lanier Reference Lanier1927), and medical treatments and clinical research (Goldberg and Phillips Reference Goldberg and Phillips1992). The term has a long-standing usage to denote the limits between trade unions in terms of the types of work tasks that their members perform (Spyers Reference Spyers1894, 2 and 137; Watson Reference Watson1938; Lambert Reference Lambert1991; Godfrey et al. Reference Godfrey, Jacobs and Fergus2021).
In healthcare, “demarcation” can refer to the differentiation between medical conditions, such as the distinctions between health and disease (Rudnick Reference Rudnick2000; Hofmann Reference Hofmann2021), between different mental disorders (Koi Reference Koi2022), and between depressive disease and non-pathological depressive symptoms (Tully Reference Tully2019). It can also refer to separations between biological categories, such as those between living organisms and non-living matter (Schoenmakers Reference Schoenmakers2023); between natural and unnatural states of ecosystems (Inkpen Reference Inkpen2017); between different biological species, races, and strains (Webber Reference Webber1903; Mayr Reference Mayr1996); and between cognitive and non-cognitive processes (Keijzer Reference Keijzer2021).
3. How the demarcation metaphor shapes our thoughts
The use of the word “demarcation” to denote other delimitations than those between adjacent land areas is metaphorical. The literature referred to in the previous section provides strong indications that the demarcation metaphor has contributed to shaping the concepts to which it has been applied. It has done so in at least four ways.
1. The demarcation metaphor gives the impression of a criterion that can be directly applied without any specialized knowledge or complex deliberations. If the border between two countries has been properly demarcated, then it is a trivial task to determine which of the two countries a particular spot on the ground belongs to. No expertise or special skills are required. In this respect, demarcations differ from definitions. A definition of a concept may very well refer to criteria that it takes considerable adeptness and specialized knowledge to apply. For instance, most of us know the definition of a bacterial disease, but it can sometimes be a challenging task even for a specialized physician to determine if a particular patient’s disease is bacterial or not. We should therefore distinguish between the act of defining a term and the act of determining in specific cases whether the criteria of the definition are satisfied.Footnote 9 The latter activity can be called diagnosis.Footnote 10 The demarcation metaphor can lead us to assume that the diagnosis task is as trivial as the determination of which side of a fence or a sightline a point on the ground is situated. This feature of demarcations can be called facility of diagnosis.
2. The demarcation metaphor induces us to reduce definitional tasks to comparative ones. If we have a criterion for determining whether some place close to the Germany–Switzerland border belongs to Germany or Switzerland, then this is not sufficient to know what land areas belong to Germany, or what land areas belong to Switzerland. Similarly, a demarcation that draws “the line between depression-as-disorder and non-pathological depressive symptoms” (Tully Reference Tully2019, 114) does not tell us how to distinguish between depression and other pathological conditions. A demarcation between natural and pathological ecosystems may not tell us how to distinguish between a natural and an artificial ecosystem (cf. Inkpen Reference Inkpen2017). This feature of demarcations can be called mere comparativeness. Some authors who have developed demarcations have shown awareness of this limitation. For instance, Engstrøm (Reference Engstrøm1996, 370) acknowledged that his demarcation between metaphorical and literal linguistic expressions does not tell us how to distinguish between metaphorical expressions and other nonliteral expressions.Footnote 11
Most of the demarcations proposed in the literature follow this pattern. They propose a criterion for distinguishing between two categories without providing a full definition of either of the two categories. However, there are also a few cases of what we can call “monadic demarcation”, namely the demarcation of a single concept, such as the concept of cognition or that of an ethical problem. These demarcations can of course be constructed as comparisons between a concept and its negation, in these cases: cognition versus non-cognitive processes and ethical problems versus non-ethical problems. However, it turns out that even in these cases, the demarcation metaphor has an impact on how the conceptual investigation is performed. This comes out clearly in an article on “the demarcation of the ethics of care”, in which the authors strive to “demarcate the boundaries of an ethics of care”, thereby “mapping out the boundaries with other disciplines” (Klaver et al. Reference Klaver, van Elst and Baart2014, 756). The focus on finding the boundaries of a discipline, rather than, for instance, its core can have consequences for the outcome of a conceptual analysis. In an interesting passage in an article on the demarcation of cognition, Fred Keijzer claims that the traditional “focus on necessary, let alone sufficient conditions” is misguided, and argues that it is better to “simply study the cases that are present in the world and work from there” (Keijzer Reference Keijzer2021, S137). It should be noted, though, that a definition based on known cases rather than general principles runs the risk of ephemerality due to difficulties in dealing with new phenomena.
A particularly elaborate attempt at defining by means of demarcation can be found in an article by Alexander Bain (Reference Bain1888). He set out to define four academic disciplines, namely psychology, logic, ethics, and philosophy. More precisely, he wanted to “to ascertain the best mode of distributing the materials, so as to be able to say of any fact or doctrine that its suitable place is in one rather than in another” (Bain Reference Bain1888, 527). His chosen method was to identify the “line of demarcation” between each two of them (but he excluded the two combinations logic–ethics and philosophy–ethics, presumably because he considered these demarcations to be trivial; Bain Reference Bain1888, 529). However, his method was obviously flawed. Pairwise demarcations do not sum up to full definitions.
3. The demarcation metaphor is strictly binary in the sense of suggesting two mutually exclusive alternatives with no overlaps. Geopolitical boundaries are conceived as “infinitely thin” (Smith Reference Smith and Tambassi2019, 148). Along the border between Norway and Sweden there is no intermediate “no man’s land.” Neither is there any piece of land that is both Norwegian and Swedish, nor any land that is partially Norwegian and partially Swedish. When we use the demarcation metaphor in other contexts, it can induce us to think in terms of a similar, strictly binary, structure. This is the feature of strict binariness. In cases with a fuzzy border, for instance between sapwood and heartwood or between a gangrene and adjacent normal tissue, it is often said that there is no line of demarcation (Eades Reference Eades1937, 470; Garlock Reference Garlock1931, 1103).
In many of the applications of the demarcation metaphor mentioned in Section 2, there is a tension between the strict binariness of the metaphor and the properties of the concepts it is applied to. For instance, a strict line of demarcation between two subjects, such as physics and chemistry, seems implausible. We would rather expect to find an overlap between the two disciplines. Similarly, a strict line between different mental disorders seems implausible since some persons satisfy the diagnostic criteria of more than one disease. Other demarcations, such as that between metaphorical and literal statements may have the opposite problem, namely that some statements seem to fall between the two categories.
4. The demarcation metaphor imparts the impression of reasonably permanent criteria. At least in peaceful times, borders between countries change very seldom. Therefore, the notion of a demarcation carries the connotation of something that is unchanged over protracted periods of time. This can stimulate a tendency to presume that what is on one side of the limit at one point in time will not move to the other side at some later juncture. We can use the term permanence for this presumption. It can be problematic if the objects of demarcation change over time. This applies, for instance, to scientific disciplines like chemistry and physics. They have “fairly sharp but time-fluid boundaries” (Hiebert Reference Hiebert1996, 94), and a demarcation between them will have to follow their developments.
4. Demarcating science and pseudoscience
Let us now turn to the application of the demarcation metaphor to the distinction between science and pseudoscience, which was introduced by Karl Popper and is still the standard terminology in philosophical discussions on this distinction. In the previous section we identified four ways in which the use of this metaphor can shape our understanding of the concepts it is applied to. We will now consider the implications that each of them can have on our conception of the science–pseudoscience distinction.
3.1 Facility of diagnosis
Pseudoscience is something that poses as scientific but is not scientific. Therefore, in order to determine whether some statement is pseudoscientific, we need to know whether it is scientific or not. Equally obviously, such determinations have to be made by scientists with the expertise required to assess the statement in question. In order to know whether a claim about chemical reactions is pseudoscientific, we need to know if the claim in question is reasonable from the viewpoint of chemical science, which is a question to be answered by chemists. The situation is similar for questions about pseudoscience in other areas. Importantly, this does not mean that the definition of pseudoscience has to be different in different areas. It can be the same for different areas, but the diagnosis will have to rely on expertise in the area(s) of science that are concerned in each particular case.
Popper’s criterion of demarcation presumes facility of diagnosis. In Logik der Forschung he emphasized that according to his criterion, the falsifiability of a claim or set of claims depends on its “logical form” (Popper Reference Popper1935, 12; Popper Reference Popper1959, 41).Footnote 12 Determining the logical form of a sentence requires competence in logic rather than in its specific subject matter or in the branch of science in which that subject matter is studied. Most of the demarcation criteria that have been proposed after Popper also presume facility of diagnosis. Larry Laudan’s (Reference Laudan, Robert and Laudan1983) attempted dismissal of the distinction between science and pseudoscience has the same problem. He claimed that there is no “epistemic” criterion that can be used for that purpose. He did not clarify what he meant by an “epistemic” criterion, but it can be seen from his examples that he referred to criteria that are easily accessible to philosophers of science, such as whether a theory makes predictions, whether it depends on ad hoc hypotheses, and what use it makes of inductive reasoning (Laudan, Reference Laudan, Robert and Laudan1983, 124). He shows no awareness of the distinction between definition and diagnosis, or of the need for qualified scientific expertise in performing the latter of these tasks.Footnote 13
3.2. Mere comparativeness
As we saw in Section 3, demarcations differ from definitions in not providing a full characterization of the concepts involved. A demarcation of pseudoscience against science tells us how to determine, for a claim that we know to be either scientific or pseudoscientific, which of these it is. There are philosophically and practically interesting questions about pseudoscience that we can answer with the help of a full definition of pseudoscience, but not if we only have a demarcation of it from science. For instance, there are many other categories of non-scientific claims than that of pseudoscience, including religious, metaphysical, and aesthetic claims (Hansson Reference Hansson and Humphreys2016). A demarcation between science and pseudoscience does not tell us how to distinguish any of these categories from pseudoscience.
Unfortunately, Popper did not seem to distinguish between demarcation and definition. To the contrary, he said that the purpose of his falsifiability criterion was “to formulate a suitable characterization of empirical science, or to define the concepts ‘empirical science’ and ‘metaphysics’ in such a way that we shall be able to say of a given system of statements whether or not its closer study is the concern of empirical science” (Popper Reference Popper1935, 9–10; Popper Reference Popper1959, 37). However, his criterion of falsifiability did not amount to a definition of either of the concepts. In order to determine whether a system of sentences belongs to empirical science we need to know more about what empirical science is than just how to distinguish it from metaphysics (or from pseudoscience).Footnote 14 Most of the discussion after Popper has continued to focus on demarcations rather than definitions of the key concepts. In consequence, philosophical issues about pseudoscience that cannot be analyzed based only on its demarcation from science have not been much discussed.
3.3. Strict binariness
As we noted in Section 3, a demarcation between the dominions of two countries allows for only two alternatives. There is no overlapping or jointly held land, and neither is there any no-man’s-land that is not fully possessed by either of the two countries. The application of the demarcation metaphor to the distinction between science and pseudoscience suggests that there is no overlap between the two categories and that there are no unclear borderline cases. Arguably, the lack of overlaps is unproblematic in this case, since pseudoscience and science are usually conceived as mutually exclusive. On the other hand, the exclusion of unclear borderline cases is quite problematic. By any definition of pseudoscience, it must involve a considerable deviation from the scientific consensus. Such deviations come in many degrees rather than as an all-or-nothing distinction. As noted by Robert Pennock (Reference Pennock2011, 183–84), Massimo Pigliucci (Reference Pigliucci, Pigliucci and Boudry2013, 13), and others, it is only to be expected that there are difficult borderline cases which cannot easily be classified as bona fide science or outright pseudoscience. By no means does this make the distinction useless. It is one of the many distinctions that we need and can use efficiently although there are cases that are difficult to decide. Other such useful but unsharp distinctions are those between morally right and morally wrong, between clarifying and misleading explanations, and between well-written and badly written texts.
Popper’s criterion of falsifiability in terms of logical form provides a strictly binary distinction. Perhaps more importantly, Laudan’s dismissal of the distinction between science and pseudoscience is based on his conclusion that there is no criterion for this distinction that has “sufficient precision that we can tell whether various activities and beliefs whose status we are investigating do or do not satisfy it; otherwise it is no better than no criterion at all” (Laudan Reference Laudan, Robert and Laudan1983, 118). As already mentioned, this is an unreasonable demand on such a criterion. The demarcation metaphor, to which Laudan referred repeatedly, may well have contributed to the surprisingly widespread assumption that a criterion to distinguish between science and pseudoscience is worthless if it does not, as Pennock puts it, provide “a precise line that can unambiguously rule any possible theory in or out of science” (Pennock Reference Pennock2011, 184).
3.4. Permanence
Since it is an essential property of pseudoscience that it is at variance with science, any reasonable definition of pseudoscience must refer to science. Science changes all the time, and therefore a statement that was scientific at one time can be pseudoscientific at another (Hansson Reference Hansson2009, 239; Boudry Reference Boudry, Pigliucci and Boudry2013, 92). Some criteria for the demarcation of pseudoscience seem to suggest stable and unchanging classifications. This applies, for instance, to Popper’s criterion of falsifiability in terms of logical form. Notably, Laudan’s dismissal of the science–pseudoscience distinction presupposes that such a distinction would have to be unchanging.Footnote 15 However, several of the most prominent proposals for criteria of demarcation recognize the advance of science as one of its essential characteristics. This applies, for instance, to demarcation criteria that refer to puzzle-solving (Kuhn Reference Kuhn and Paul1974) or to the progressivity of a research program (Lakatos Reference Lakatos, Stuart, Fauvel and Finnegan1981). It is therefore doubtful whether the demarcation metaphor has prevented scholars from taking the ever-changing nature of science into account when discussing the distinction between science and pseudoscience.
5. Conclusion
We have found strong indications that the demarcation metaphor has impacted negatively on the philosophical discussion on pseudoscience in at least three ways. (1) The metaphor gives the impression that it should be as simple and trivial to distinguish between science and pseudoscience as it is to determine on which side of a fence or a sightline a point on the ground is situated. This has led to a search for criteria that can be used without specialized scientific knowledge of the contested statements. (2) The metaphor has led to a focus on criteria for distinguishing between science and pseudoscience, rather than full definitions of the two concepts. This seems to have contributed to neglect of important philosophical issues, such as how other forms of non-science relate to pseudoscience. (3) The metaphor induces us to think in terms of a sharp line between science and pseudoscience with no borderline cases. This is unrealistic, since scientific quality comes in degrees rather than as an all-or-nothing phenomenon.
Karl Popper’s falsifiability criterion for distinguishing between science and pseudoscience has often been criticized for yielding implausible categorizations.Footnote 16 The above analysis goes further and criticizes the framework that he introduced for this discussion. My proposal is that we should give up the demarcation metaphor. Instead, we should pursue the two tasks of defining pseudoscience—a task primarily for philosophers—and diagnosing specific cases of seeming deviations from science—a task primarily for scientific experts in the concerned areas of science. This division of labor might lead some to believe that there is not much for philosophers to do in this area, but that is not so. There are several important philosophical issues connected to the definition of pseudoscience and to its general features. Many of them have as yet received very little attention.Footnote 17 Some of them are:
What are the objects of definition? Several types of objects can be classified as pseudoscientific. For instance, arguments, statements, beliefs, concepts, hypotheses, theories, doctrines, research activities, and research projects have all been described as pseudoscientific. In this cluster of related concepts, it would seem reasonable to select one of them as a primary definiendum, in terms of which the other members of the cluster are defined. Is that a feasible approach and, in that case, what should be the primary definiendum?
What are the subject areas? Pseudoscience is usually conceived as operating within topics that science can adjudicate. However, not all academic disciplines are included in science. The English term “science” differs from the German “Wissenschaft” in excluding the humanities. Holocaust denialism is clearly “Pseudowissenschaft,” but is it “pseudoscience?” In practice, the word “pseudoscience” is often used in a widened sense that corresponds better to the German “Wissenschaft” than the English “science.” Is there a principled way to distinguish between topics in which (certain types of) misconceptions can be called pseudoscience and topics in which that designation is not applicable?
Can a claim be pseudoscientific even if no one calls it science? Informally, pseudoscience is often defined as something that is falsely claimed to be scientific. But is such a claim an adequate defining characteristic? Consider two astrologers who make the same unscientific claims. One of them says that astrology is science, whereas the other says that astrology is something else that is much more reliable than science. Probably, both would be described as engaging in pseudoscience. Is that a reasonable judgment, and in that case, how can it be justified?
Can research be pseudoscientific without involving any false or unwarranted claims? There are examples of large resources being spent on searching for something that is almost sure not to exist. That is wasteful, but is it a reason to call the activity pseudoscientific? Many would call a search for the Loch Ness monster pseudoscientific, but is it so if no false claims are involved?
How does pseudoscience differ from other types of deviant or deficient science? There are many other types of bad or failed science, such as honest mistakes, scientific misconduct, scientific fraud, science denialism, and conspiracy theories. To what extent can pseudoscience overlap or coincide with each of these? Is there some specific characteristic that distinguishes pseudoscience from all other types of deviations from good science? Potential candidates are a scientific pretence, a deviant doctrine, and strategies that immunize against valid counterarguments.
It is high time to give up the pretension that philosophers are called and qualified to assess scientific quality in all areas of science. Instead, we should focus on the many truly philosophical issues that pretended, failed, and deviant science gives rise to.
Acknowledgments
The author wishes to thank the editors and anonymous referees of Philosophy of Science for most useful comments.
Funding Information
This research was not supported by any specific funding. The author has received funding from the Swedish Research Council for an epistemological project (grant 2020-01460) and from the European Union’s Horizon 2020 program for an ethics project (grant 101004594).
Competing interests
None to declare.
Open access
