Abstract
This paper presents and discusses various strategies that have been wielded against scientism, roughly the claim that only science can give us knowledge. The strategies identified are: (1) the counter example strategy, (2) the denying of claimed entailments of science strategy, (3) the self-undermining strategy, (4) the presupposition strategy, and (5) the limits of science strategy. In addition, two proposals are discussed that aim to recast the debate about scientism in a way that renders these strategies obsolete. It is argued that these proposals are misguided.
Keywords
Scientism is a claim. Or, better, it is the name of a claim. Or, even better yet, it is a name that has been given to a variety of claims about the natural sciences. Until recently, the name was a term of abuse, and used to belong to the family of terms of which ‘fundamentalist’ and ‘racist’ are other members – terms that are other-applied, not self-applied: it is someone else who is a fundamentalist, a racist, or a ‘scientismist’ ( = friend of scientism), not oneself. The reason for this is that these terms come with, or even instantiate, condemnation. And people normally don't condemn their own views but, when it seems warranted, they do want to condemn the views of others. With respect to ‘scientism’, things are beginning to change, however. Various authors now self-apply that term, not to condemn themselves, but as a badge of honour. Examples include Alex Rosenberg (Rosenberg 2011, 2018), James Ladyman, and Don Ross (Ladyman and Ross 2010). The honour that the badge intends to signal is this: being a scientismist is to be fully serious about science, to be fearless about it, to follow it wherever it leads. There is a parallel with the term Impressionism, that was once used to criticize and insult painters like Monet, Renoir, Degas, and Cézanne, but was later on appropriated by these artists themselves as a badge of honour. I note that currently Impressionism is mainly used as a descriptive term without condemnatory moral overtones.
In this paper, I use scientism analogously to how Impressionism is used nowadays, so mainly as a descriptive term, shorn of condemnatory overtones. Scientism, purely descriptively understood, is, as I indicated, the name for a number of claims about science. Here is one claim that has been graced with the name:
Scientism … is the conviction that the methods of science are the only reliable ways to secure knowledge of anything; that science's description of the world is correct in its fundamentals; and that when ‘complete’, what science tells us will not be surprisingly different from what it tells us today … Science provides all the significant truths about reality, and knowing such truths is what real understanding is all about. (Rosenberg 2011, 6–7)
The exact meaning of this claim depends on what ‘science’ refers to. There is a rather traditional sense of science (scientia) according to which all sorts of organized or well-founded bodies of knowledge or skills were called a science. Being a science had nothing to do with subject matter. John Locke had this notion in mind when he claimed that physics and chemistry, or, as he called it, ‘natural philosophy’, are not capable of being made into a science (Wolterstorff 1996, 28–37), whereas ethics is capable of being made into a science. Hobbes regarded politics as one of two perfect examples of a science; the other being mathematics (Hobbes 1656, 183–4). Usage of this notion of science lingered on well into the twentieth century. In 1935, for example, a book with the title Approaches to a Science of English Verse (Schramm 1935), and in 1960 a theological book with the title An Introduction to the Science of Missions (Bavinck 1960) could appear without it being felt that the word science was misapplied.
The author of the description of scientism that I just quoted, Alexander Rosenberg, had a rather different notion of science in mind when he penned it down. That poetry or evangelistic missions could be proper subject matters for science was very far from his mind indeed. For him science concerns certain subject matters, and not merely the systematicity in which whatever subject matter there is can be described. Among others, he had in mind physics, chemistry, evolutionary biology, and brain science. He doesn't give a description of what it is for something to qualify as a science, nor does he say what he takes knowledge to be – but he is relying on widely shared ideas about science, such as that (i) science is empirical and (ii) scientific knowledge is knowledge that has been put to very rigorous tests.
Rosenberg's description of scientism has an element that merits special attention, viz. that it includes the idea that the current scientific description of the world is correct in its fundamentals. I take this to mean that new breakthroughs are not to be expected, and that the future for science is filling in details within an overall frame that is fixed in its fundamentals. Rosenberg, then, adopts
And he also adopts what I call
It should be clear that the latter entails the former, but the former not the latter. There are other claims that have also been named (forms of) scientism (Stenmark 2001). One that shall figure in my discussion is:
As to where science is ‘correct in its fundamentals’, Rosenberg claims that science has established all of the following (Rosenberg 2011, 2–3):
E1 There is no God. E2 The nature of reality is what physics tells us that it is. E3 There is no purpose of the universe. E4 Life is meaningless. E5 There are no souls (substance dualism is false). E6 There are no persons. E7 Free will is an illusion. E8 There is no life after death. E9 There is no moral difference between (what we call) right and wrong. E10 Concerning abortion, euthanasia, suicide, paying taxes, foreign aid, anything goes; there are no moral obligations, no moral permission, and no moral prohibitions. E11 There are no beliefs (this thesis is usually named ‘eliminativism’) (Rosenberg 2018).
Although this list cries out for comments, I must forego this, and rely on an intuitive understanding of the claims. But even such an understanding indicates why scientism merits critical scrutiny. For if theses E1–E11 were to be true, that would be astonishing and heart-breaking news, as it would undermine many of our most cherished beliefs and assumptions. After all, we believe we are persons, but according to Rosenberg there are no persons. And we believe that rape, torturing children, and discrimination of people of colour are deeply morally wrong, but according to Rosenberg the difference between what is morally right and morally wrong is illusory. We also believe that we have various moral obligations, and that there are moral prohibitions, but Rosenberg denies all of this. And we believe that we have beliefs, but Rosenberg denies that too. And so it goes for many of the other entries on the list. Many people believe them, they are even crucial to their self-understanding and understanding of the world, but according to Rosenberg they are all wrongheaded. A view that is alleged to have these implications merits close critical scrutiny (Van Woudenberg, Peels, and de Ridder 2018).
My aim for this paper is to map the moves that a critic of scientism can make such that, when they can successfully be made, scientism is rendered implausible, if not falsified. In the sections that follow, I identify and discuss five anti-scientistic argumentative strategies. What I describe are strategies as opposed to fully worked-out arguments. Each of the strategies can be elaborated upon, and tweaked in various directions, in ways that are analogous to how the military strategy of a fifth column can be used: by a Trojan horse, by a slow settlement of, say, ever more Russians in Marioepol, or by a delegation of apparent tourists that are visiting the city. My focus is on what the strategies are, not on the particulars of a specific execution.
This paper is organized as follows: sections 1–5 describe the argumentative strategies against scientism. The use of these strategies has been criticized. In section 6, I look into these criticisms only to find them misplaced. The final section takes up the question of what we can learn from the study of scientism.
The counterexamples strategy
The first argumentative strategy against scientism, understood as Strong Scientism, that can be found in the literature is to offer what seem to be counterexamples (Van Woudenberg, 2018a; Van Woudenberg and Rothuizen-van der Steen 2016; De Ridder 2018). That is to say, the strategy is to offer examples of things that we know but not on the basis of science. There would seem to be many such examples. For instance, I know my name, I know that currently the sun is shining, I know that many dogs bark, I know that I ought to keep my promises, I know that I have a body, I know what I am currently thinking about, I know that I can quote poems from Robert Frost, I know that I own some guitars. And this is nothing special about me. Virtually everybody knows humble things similar to the humble things that I know. And if we have no reason to doubt this, then, so this strategy tells us, we have a good reason to reject Strong Scientism.
Friends of Strong Scientism, if they want to maintain their position, should either argue that we do know these things through science after all, or deny us the humble knowledge that we have. The first response seems a non-starter. For my knowing my name, and you yours, isn't acquired by anything that qualifies as science, or as scientific investigation. Likewise, perceptual knowledge that the sun is now shining isn't acquired through science (Van Woudenberg and Rothuizen-van der Steen 2016). And so on, and so forth. The other alternative isn't very promising either. How could it be argued that I don't ‘really’ know my name? How could it be argued that at this moment I don't ‘really’ know that the sun is shining? Etc. How could it be argued that you don't really know the similarly humble things that you think you know?
Well, there are, of course, sceptical arguments, such as Peter Unger's, that explicitly purport to show that we don't really know these humble things (Unger 1974). Such arguments, however, are to no avail in defence of Strong Scientism. For if they are really successful, then we have no knowledge at all, and a fortiori no scientific knowledge. So, epistemological scepticism doesn't rescue Strong Scientism.
Now you might think that if Strong Scientism succumbs so easily to the counterexample strategy, then there isn't much value in paying so much attention to it. This is what Hietanen and colleagues say (Hietanen et al. 2020). Strong Scientism is too easy to refute, and too implausible to merit serious discussion. I disagree. First, as Jeroen de Ridder has shown (De Ridder 2014), many popular science books implicitly or explicitly adopt or push something like this position, and many of those books are widely read and hence have impact. Not to pay attention to it because of its implausibility or faultiness compares to not rebutting fake news because those who are well-informed know it is utter foolishness. And second, among philosophers Strong Scientism is only explicitly advocated by Rosenberg, and perhaps also by Ladyman and Ross. Yet that is no reason to devote no serious attention to it. There is, I suggest, a comparison here with the study of radical scepticism. As many epistemologists have said, there are only very few true radical sceptics, and yet it is worthwhile to study sceptical arguments, because we can learn something from it about knowledge (Greco 2000, 2–3). Likewise, studying Strong Scientism pays off, because we can learn something from it about knowledge, and knowledge acquisition, as well as about sources of knowledge. In the final section of this paper, I return to this.
Given the strength of the counterexample strategy, friends of scientism can revert to Weak Scientism, i.e. the view that the methods of science are the best ways to secure knowledge of anything. But this view is problematic too. If something is the best way, there should be an alternative compared to which the best way is better. The general strategy against Strong Scientism can, with some minor modifications, be wielded against Weak Scientism as well. All we have to do is to provide examples of things we know, and then ask whether scientific research delivers something that is epistemically better than something that does not qualify as scientific research. Such examples are widely available. I know that the sun is currently shining, as I see it with my own eyes. Suppose now that the local meteorological institute (that heeds all the canons of scientific prediction) broadcasts the weather report that includes the news that where I live the sun is currently shining. Is believing on the basis of the weather report better than believing on the basis of my first-hand experience? It would seem not. For, if I see with my own eyes that the sun is shining, and the meteorological institute issues a report that says that it is overcast, then it is entirely reasonable to go by my own experience, and neglect the science-based weather report. Sometimes first-hand experience trumps science-based belief (Ranalli 2021). Or take this case: I know that the hands I use to type this sentence belong to a body that is my body. I know this through an acquaintance with one specific body, my body, that I have with no other body. Can I come to know through scientific research that the body to which the hands that type this sentence is attached is my body in a better way? What would that research look like? That is not so clear. But then why think that science is the best way to obtain that knowledge? As a final example, consider my knowledge that if I make a promise, I thereby incur the obligation to do as promised. I know this, perhaps through a moral sense, perhaps through moral intuition, perhaps through some kind of reflection – but the fact is that I know this. But research in physics, chemistry, biology, and other natural sciences seems unable to give me that knowledge. So, it isn't true that in each and every case of knowledge science provides us with the best avenue towards it. The latter example even suggests that science is no avenue to moral knowledge at all. This means that Weak Scientism is as problematic as her Strong sister. The strategy of counterexamples is equally successful against both.
The denying the claimed entailments of science strategy
If the argument against Strong Scientism is a good one, then it is a good one against Super Strong Scientism as well, as the latter entails the former. But there is another argumentative strategy against Super Strong Scientism – viz. denying that science has established E1–E10, so denying that science has established that there is no God, that there is no purpose in the universe, that life is meaningless, that free will is an illusion, and all the rest. (Remember that Super Strong Scientism claims that science is correct in its essentials and that E1–E11 follow from these essentials.)
Is this strategy successful? This is, of course, a gigantic question that it is impossible to materially enter into. What I will do, therefore, is just flag the fact that many well-informed and intelligent scientists, scholars and philosophers have concluded that science just does not show this.
Does science entail that there is no God? This idea is widespread, but for various reasons problematic. First, historically speaking the enterprise of modern science was started mainly by faithful Christians (Harrison 2015), and many scientists now living are theists (Ecklund 2019). They, at least, didn't and don't think: ‘This is what we discovered through scientific inquiry, hence there is no God.’ Second, how is science supposed to show that there is no God? Does the discovery of laws and regularities in nature, or the discovery of cells, or the discovery of DNA, or the idea of evolution, or of the Big Bang entail so much? It has been argued by many that this just isn't the case (Stewart 2010). Third, there are strong philosophical arguments for the existence of God (Swinburne 1991; Craig and Moreland 2012). My point here is that many have argued that science soberly taken just doesn't entail the nonexistence of God.
Since I shall deal with E2 in the next section, I move on to E3 and E4, that there is no purpose in the universe and that life is meaningless. Often, these claims have to do with the role that chance and randomness play in scientific theories, such as evolutionary theory and quantum mechanics. But many scientists and philosophers who have given these claims serious thought have come to the conclusion that science just doesn't show that there is no purpose of the universe or that life is meaningless (Klapwijk 2008; Plantinga 2010; Alexander 2018; Van Woudenberg 2014). Perhaps these claims are true. But if they are, science doesn't show it.
Is it an entailment of science that there are no souls, and that substance dualism is false? Again, this has been contested by well-informed philosophers like E.J. Lowe (Lowe 1996), Alvin Plantinga (Plantinga 2007), and Richard Swinburne (Swinburne 2019). Of course, there are also philosophers who deny the existence of souls, and who criticise substance dualism. What this suggests is that E5 is the matter of a philosophical discussion, and that, regardless of how one comes down on the issue, it is underdetermined by scientific inquiry.
Since E6 (and E10) will figure in a later section, I move on to E7, the claim that free will is an illusion. There are various theories about free will: compatibilist theories (according to which free will is compatible with determinism), as well as incompatibilist ones (according to which free will and determinism are incompatible); and one can be an incompatibilist by denying free will (which is what hard determinists do), but also by denying determinism (which is what libertarians do). Again, it has been argued that the relevant science doesn't even show that libertarian accounts of free will are false (O'Connor 2005, 2020; Swinburne 2013; Jongepier and Cassam 2020).
As to E7: does science show that there is no life after death? Again, it has been argued that life after death is compatible with all that science tells us about humans (Zimmerman 2012; Swinburne 2019; Ducasse 1961).
Finally, entailments E8 and E9, that there is no moral difference between (what we call) right and wrong, that there are no moral obligations, no moral permission, and no moral prohibitions. This too is a very large subject matter. Many who issue these claims do so on the basis of evolutionary accounts of morality. But again, that science, especially evolutionary science, shows that morality is an illusion can be and has been contested on various grounds (Clayton and Schloss 2004).
If Super Strong Scientism claims that science shows that E1, E3, E4, E5, E7, E8, and E9 are true, while that claim is spurious at best, so this strategy goes, then there is excellent reason to reject Super Strong Scientism (Arbesman 2013).
The self-undermining strategy
Another argumentative strategy against scientism is to attempt to show that it is self-undermining (Ratzsch 1986; Van Woudenberg 2012; De Ridder 2014; Peels 2020). The strategy goes like this: Strong Scientism is the claim that only science can give us knowledge. But if we look at what science gives us, we just don't see that it gives us Strong Scientism. Spelled out a bit further: science tells us many things, many things that we have even come to know through scientific research. We have come to know, through scientific research, facts about superconductivity at extremely low temperatures, about hereditary relations between organisms, about the causes of cancer, and about a zillion other things. But we haven't come to know through scientific research that Strong Scientism is true. We haven't come to know through scientific inquiry that only science can give us knowledge. Strong Scientism is not a deliverance of science. This means that we do not know that Strong Scientism is true, and also that supporters of Strong Scientism don't know that Strong Scientism is true. But then friends of Strong Scientism should not assert Strong Scientism – for there seems to be a rule of assertion according to which we shouldn't assert what we don't know (Williamson 2000, 238–69). But friends of Strong Scientism do assert Strong Scientism. Hence, they do something they shouldn't do. They do something that, given the rule of assertion, is self-undermining.
There is another way to wield the self-undermining strategy against Strong Scientism. This can best be shown when we formulate Strong Scientism in a slightly different way, namely as the injunction to only accept the deliverances of science. But, as intimated above, Strong Scientism itself is not a deliverance of science. This means that someone who accepts Strong Scientism has an excellent reason to reject Strong Scientism. The position undermines itself, it is self-referentially incoherent (Mavrodes 1985).
Those who accept Strong Scientism, then, don't accept it on the basis of science, but on some other basis, perhaps on the basis of allegiance to some form of naturalism. But naturalism is no deliverance of science either.
So the strategy is to argue that if one accepts Strong Scientism, one has a good reason to reject it, because scientism is not a deliverance of science.
The self-undermining argument can also be used against Weak Scientism. Weak Scientism, recall, tells us that the methods of science are the best ways to secure knowledge of anything. Suppose now that we want to know what we can know. Then Weak Scientism says that the best way towards an answer is to use the methods of science, say, the methods used in Relativity Theory, QM, evolutionary biology, and so on. Applying these methods, says Weak Scientism, should give us the best answer to our question. However, these sciences inform us about the physical world, and about biological life, and much else besides. But application of these methods doesn't seem to answer our epistemological question as to what we can know at all. Let alone that those methods give us the best answer to it. Weak Scientism, then, is no deliverance of science either. But if it isn't, then those who accept Weak Scientism have an excellent reason to reject it. The position undermines itself. Another way to put the point is this: those who accept Weak Scientism can't know that Weak Scientism is true. Or, more precisely, those who accept Weak Scientism cannot claim that the methods of science are the best way to answer the question of what we can know. This means that those who accept Weak Scientism don't do so on scientific grounds. They accept it on some other basis or on no basis. But, if they do, they have in fact abandoned Weak Scientism (Ladyman 2018).
Of course, nothing in the self-undermining strategy denies that, in many cases, following the canons of science often is the best way to secure knowledge of something. But the strategy does deny that it is always the best way to secure knowledge of anything – whatever it is.
The presupposition strategy
Yet another strategy to undermine scientism is to argue that, in order for science to get off the ground, we must presuppose theses that, as the term suggests, are presupposed and not established by science (Ratzsch, 1986; Van Woudenberg 2012, 135–7). But what is a presupposition? We can distinguish at least four notions of ‘presupposition’. Here is a way to represent them:
P presupposes Q when the following is the case:
if Q is false, then P is false as well (I call this an a-presupposition) if Q is believed to be false, then belief in, or acceptance of, P would no longer be justified or warranted (a b-presupposition) if Q is (known to be) false, then P is neither true nor false (a c-presupposition) if Q is believed to be false, it doesn't make sense to engage in activity P (a d-presupposition)
Note that Q is always a statement, whereas P is a statement in (a), (b), and (c) or an activity in (d). I offer an example of each type. The statement ‘John was at home this morning’ a-presupposes the statement that ‘Somebody was at home this morning’. For if the latter statement is false, then so is the former.
If someone bases his belief that ‘Susan is an insecure person’ on the outcome of a Rorschach test, the statement that ‘the Rorschach test is a reliable test for personality traits’ is a b-presupposition of that person's belief. For if the latter statement is believed to be false, belief in the former proposition is no longer justified or warranted.
‘The present king of France is bald’ c-presupposes ‘Presently France has a king’, to use a famous example. If the latter is false, then the former is neither true or false. If the former is to be either true or false, the latter must be true.
The activity of playing tennis has as a d-presupposition the statement that balls can be hit by means of rackets. For it is, somehow, incoherent to play tennis and yet explicitly deny that balls can be hit by rackets.
The strategy under discussion is to argue that science has presuppositions. It has been argued that science has metaphysical, epistemological, and axiological presuppositions (Van Woudenberg 2012). Metaphysical presuppositions include (1) that the world displays order, and (2) that truths about the physical world exist independently of their being known or believed by humans. Epistemological presuppositions include (3) that human beings are capable of knowledge, and (4) that the faculties we use in doing science are not massively unreliable. Axiological presuppositions include (5) that in doing science we ought to seek the significant, not the trivial, (6) that we ought to comply to a number of ethical rules of scientific conduct, and (7) that theorizing itself is and ought to be guided by certain norms.
Within the confines of this paper, I can only showcase the strategy under review that proceeds from (2) and (7).
Truths about the physical world exist independently of any human being believing them, knowing them, affirming them, or even entertaining them. Such truths exist mind-independently. The truths we seek in science are truths that already exist ‘out there’ prior to our getting to know them through scientific investigation (Searle 1995). And a statement S is true if and only if the world is the way S says it to be. The statement that ‘Lemons are sour’ is true if and only if lemons are sour; nothing more is needed, and nothing less will suffice (Alston 1996). Of course, in order for us to know truths, we, knowers, must exist. But in order for the truths to exist, our existence isn't needed.
To see this, let us engage in a little thought experiment. Suppose that through some unhappy course of events, in the year 2050 all human beings have gone extinct. Perhaps an atomic rain has extinguished all human life on earth. Now concentrate on the world in 2050, and ask yourself: ‘Would there still be truths in the year 2050?’ Then the correct answer would seem to be: ‘Yes, of course! It would still be true that water evaporates in an open container, and that two heavy objects attract each other with a force that is directly proportional to the product of their masses, and inversely proportional to the square of the distance between them, etc. etc.’ Of course, in 2050 none of these truths would be known by anybody (except God and other possibly non-human intelligent beings). But this fact doesn't rob these truths of their very existence. The annihilation of humans doesn't bring with it the annihilation of all truths.
When doing science, it must be presupposed that truths exist independent of our believing or knowing them. Unless we assume that there is a truth about what causes smallpox, scientific research into the causes of smallpox has no clear aim. Unless we assume that there is a truth about whether a physically active lifestyle is healthy or not, scientific investigation into the health effects of regular physical activity is pointless. Of course, prior to scientific inquiry we don't know what the truth about these things is. But we do and must assume that these truths exist.
That truths exist is, therefore, a presupposition of science, not a result thereof. It is a d-presupposition, for you cannot coherently engage in scientific research and at the same time deny that truth exists. It is, moreover, also a b-presupposition, for if truth doesn't exist, belief in a scientific proposition is pointless. The reason for this is that to believe a proposition just is to believe the proposition to be true. It just doesn't make sense to say ‘I believe proposition P to be true, but truth doesn't exist’. And if truth doesn't exist, belief in a scientific proposition can never be justified – at least, this is so on the plausible assumption that justification and truth are related in this way: that the more and better justification one has for believing something, the more likely it is that the belief will be true.
As to (7), the argumentative strategy is again to argue that science presupposes the existence of certain values, values that we haven't come to know through scientific research. In science, we give reasons for and reasons against hypotheses and theories. Some reasons for a particular theory are good, sound, or valid, whereas others are not. In order to devise and evaluate scientific hypotheses and theories, we hence must be aware of norms that distinguish good and bad reasonhood. It is no small task to specify these norms. For present purposes, it will suffice to just mention ‘simplicity’ (Swinburne 1997; Laudan 1986). That theory A is simpler than theory B, while both explain the data equally well, is a good reason to prefer A over B. One theory is ‘simpler’ than another in the relevant sense, other things being equal, when the one poses fewer entities, or fewer kinds of entities, or fewer relations, or fewer kinds of relations than the other. A theory that explains all the evidence but refers to only one killer is to be preferred over another that explains the evidence equally well, but refers to a gang of killers. We presuppose that, ceteris paribus, simpler theories are to be preferred over more complex ones.
The strategy here, then, is that this, or any other, norm for good reasonhood is a presupposition of science, not something that has been established by science. It is not an a- or a b- but a d-presupposition. For it is incoherent to engage in scientific research and yet deny simplicity or any other norm for good reasonhood. Science is impossible without views on good reasonhood.
This strategy targets both Strong and Weak Scientism, as science figures in both of them, and science, so the strategy argues, cannot come off the ground without making certain assumptions.
The limits of science strategy
Another argumentative strategy is to argue that there are general limitations on science, by which is usually meant that there are certain things that science is unable to deliver, and hence that there are things we should not expect science to deliver (Ratzsch 1986; Rescher 1999; Medawar 1984; Moreland 1992, 103–38). This strategy is a close ally of the counterexample strategy. Analogies can hammer home the general intent of the strategy: there is this particular limitation to a guitar that it cannot be used as an equipment to fly across the Atlantic; there is this particular limitation to a thermometer that it cannot measure wind speed; there is this limitation to quantum mechanics that it cannot tell you how to dance the tango. This is, of course, no criticism of guitars, thermometers, and quantum mechanics. Likewise, the strategy of pointing out limitations to what science can deliver cannot and should not, for that reason alone, be seen as science criticism.
There are many things that science seems unable to deliver. Put this way, pinpointing limitations is easy. Science cannot deliver original Van Gogh paintings, legal constitutions, a border between East and West Germany, or a Nobel Prize winning poet, or a square circle. But then again, nobody would expect any of these to be among the deliverances of science. We expect the deliverances of science to be of a cognitive nature: knowledge, understanding, true belief, explanations, predictions. Hence, the strategy under scrutiny pinpoints cognitive limitations to science. But even this is easy. For there are many things that so far science just hasn't told us. The evidence abounds: across the globe there are researchers teamed up in research projects to figure out things that are currently unknown. And no scientismist would feel exercised. So the strategy under scrutiny is only really convincing if it can pinpoint cognitive items that science cannot deliver as a matter of principle.
One way to concretise the strategy is to focus on questions that science is seemingly unable to answer. Here is a sampling of such questions:
Q1 what does chocolate taste like? Q2 do I enjoy eating chocolate? Q3 is breaking a promise always bad? Q4 is the existence of evil compatible with the existence of God? Q5 Does time move – and if it does, with what speed? Q6 Are there propositions that are necessarily true? Q7 Why is there something rather than nothing? Q8 When is a life well-lived? Q9 Is there a meaning to earthly existence?
The natural sciences seem incompetent to answer these questions in any satisfactory way. There don't seem to be empirical methods the application of which could be expected to deliver answers. A full execution of this strategy would add detailed considerations as to why science cannot be expected to take on these questions.
A number of the questions on the list, although science cannot answer them, do seem to have decent answers nonetheless. We even seem to know some of them. This shows that the limits of science strategy, when successful, undermines Strong Scientism. For if the case can be made that science (as we now know it, and as we expect it to develop in the future) cannot give us moral knowledge, or knowledge by acquaintance, or knowledge of one's own mental states, or of modality, etc. and we presume that we have knowledge of these sorts, then science, or the methods of science, cannot be viewed as the only reliable way to secure knowledge of anything. There are other ways.
One way to make this point in more detail is by invoking the idea that there are distinct domains or fields of inquiry, and that distinct domains require distinct methods and procedures, and will yield distinct kinds of cognitive goods. For example, it has been argued that the objects that are studied in the natural sciences are distinct from the objects that are studied in the humanities. The objects of humanistic study display such features as being the product of intentional design, embodying value, and having meaning, which the objects of the sciences are lacking. Since intentions, meanings (Van Woudenberg 2021, 173–6), and values cannot be studied by methods used in the natural sciences, but can be studied by methods in use in the humanities, so the argument goes, we should guard against illegitimate annexations and hostile takeovers (Van Woudenberg, 2018b).
This strategy also cuts against Weak Scientism, at least if the case could be made that science is unable to give us better knowledge of qualia than common sense does; or better moral knowledge than our moral intuition or our moral reflection can; or better knowledge by acquaintance, or better knowledge of our own mental states than self-observation can, etc. In general, this strategy aims to undermine Weak Scientism in that it argues that it isn't plausible to suppose that science can eventually give us answers to the questions mentioned, answers moreover that are better than the answers that common sense and other non-scientific sources yield (Peels and van Woudenberg, 2021).
Meta criticisms of the scientism debate
In the previous sections, I have given an overview of argumentative strategies against Strong as well as against Weak Scientism. Recently, the complaint has been raised that these strategies are somehow ill-conceived. In this section, I pay attention to two such complaints.
First, there is the complaint by Anthony Gottlieb (2019), who calls for a ‘recasting’ of the scientism debate. He says
[P]erhaps reflecting on the way that the development of the concept of science has been intertwined with changing ideas about the best type of knowledge can free up our thinking about scientism. It may encourage us to look askance at the notion of intrinsic limits to the sciences. For if science is just what we call the best kind of theoretical knowledge, and physics, chemistry, and biology are paradigms of it because of their rigor, not because of their subject matter, then what sense is there in supposing that ‘scientific’ methods can take us only so far? (107)
If I understand Gottlieb well, he says that if we think of ‘science’ as something that is characterized by its rigour, then there is no reason to think that only physics, chemistry, and biology will be considered science. Rigorous reflection on poetry could then lead to a ‘science of verse’, as Schramm called it. And rigorous reflection on the Christian revelation could then lead to the science of God. For if science is characterized by rigour, then many areas of study that most of us aren't inclined to label ‘scientific’ will be scientific after all. To me it seems, however, that if the proposal is to recast the debate about scientism this way, scientismists would reject the recasting, as it opens the door to things they don't want.
The quotation just offered speaks of ‘theoretical knowledge’ only, which I suppose contrasts with ‘practical’ or moral knowledge. However, if science is characterised by rigour, then there is no reason to suppose that morality cannot be or become scientific too. After all, there is rigour to the various proposals to unify the seemingly wide variety of moral concepts and ideas, such as consequentialism and deontologism. Then Locke's ideal of a science of morality may come true after all. The problem with this way of recasting the debate about scientism is that proponents of Strong Scientism certainly do think this cannot be. Rosenberg, we saw, denies the existence of moral differences. And, whereas Gottlieb thinks that most critics of scientism don't take science seriously enough, I presume that Rosenberg will pay Gottlieb the same compliment.
But whatever we should think about Gottlieb's proposal about the definition of science, for the purpose of the present paper I should like to be very clear about the following. The scientism that I had in mind, and against which the argumentative strategies that I have referenced are directed, is Scientism that does assign the natural sciences a truly unique and exclusive place in the human search for knowledge. Perhaps Gottlieb thinks this position is not worthwhile discussing and that we better spend time on discussing more plausible versions of scientism. He is free to think so, of course. But I happen to disagree. By my lights, Strong Scientism is a position worthy to discuss, just as radical scepticism is.
The second complaint about how the debate about scientism is developing, issued by Hietanen et al., exudes a similar atmosphere. They discuss criticisms of scientism by Rik Peels, Jeroen de Ridder, and to a lesser extent also by myself, and their basic point is that these criticisms only target ‘strong narrow scientism’ – the view that the natural sciences are the only sources of knowledge (and justification, rational beliefs, or the like), but that they leave ‘strong broad scientism’, ‘weak narrow scientism’, and ‘weak broad scientism’ unscathed. Our way of going about, as their paper title has it, exemplifies ‘How Not to Criticise Scientism’. The problem, they say, is that we attack uncharitable definitions of scientism, like Strong Narrow Scientism. But that the weaker forms, that are much more plausible, are not targeted.
By way of reply, I make two points. Strong Scientism is not an uncharitable reconstruction of scientism. Rosenberg, as we noted, explicitly endorses the view. And it was that view that I, but also Peels and De Ridder, directed our criticisms at. I am happy to note that Hietanen and colleagues agree with Peels, de Ridder, and myself that strong narrow versions of scientism ‘definitely need to be criticized for their strong claims’ (Hietanen et al. 2020, 538) and they seem to agree with the arguments. Their point is merely that these arguments aren't successful against weaker versions. In the previous sections, I addressed this when I discussed the narrow/weak version.
My second point is that, although everybody is free to define labels the way they want, I don't think that the broad scientisms are deserving of the name. These positions, especially the weak/broad ones, are much more plausible than the narrow ones. But even these have untoward features. For in Hietanen cum suis’ scientism, pure ‘pure intuition’ is held to be unreliable. And this seems to put mathematics, that lives off pure intuition, in disregard – which is very problematic, to say the least. Their scientism also includes the claim that divine revelation is unreliable. But this too seems problematic. For, as John Locke already said, if there is divine revelation, then it should be deemed utterly reliable. The question is, whether there is (or has been) revelation – and this is something that must be argued for not on the basis of revelation, but on the basis of historical and other kinds of normal evidence (Swinburne 2008, 107–31).
What can we learn from the study of scientism?
I finally want to take up the question of what the point is of studying scientism in any of its varieties? Earlier on, I said that we can learn something from such a study, just as we can learn something from the study of radical skepticism. The aim of this section is to specify in some detail what we can learn from it.
A first thing we can learn from such a study is that scientism is a view that, for all its reverence for science, it itself is not based on scientific evidence. It is a view that is, so to say, underdetermined by scientific evidence.
Second, we can also learn from that study that it is a grave mistake to identify ‘knowledge’ (and related notions such as ‘rational belief’) with ‘whatever it is that secure natural science can and does tell us’. For first, there is much that we know without science or scientific inquiry lending a helping hand, and even without science being capable of lending a helping hand. I am thinking here of moral knowledge, as well as knowledge of common-sense propositions. Second, as the history of science shows, many deliverances of science have had to be given up over time, either because they were proven false, or they were shown to be not properly evidenced. But whatever the reason, this shows that not all the deliverances of science qualify as knowledge. In this way, then, the study of scientism teaches us not to identify ‘knowledge’ with ‘whatever it is that natural science can and does tell us’.
Third and related, we can learn that there are other sources of knowledge than scientific research. Even if working out the full details is a job of considerable magnitude (Van Woudenberg 2021), this is a plausible idea. For science, or scientific research, itself is a source of knowledge that involves, or is based on, other and often more elementary sources. There can be no science without the use of such sources as perception (which helps us to observe phenomena), reasoning (which enables us to get from what we perceive to conclusions that are far more general), reason (that gives us knowledge of self-evident truths), memory (by which we remember, among multiple other things, what experiments we are conducting, what we are trying to prove or disprove, etc.), reading and testimony (by which we come to know what others have been working on and of what they claim to have come to know through it). The point is, if we think (as we should) that science is a source of knowledge, this brings with it a commitment to the thought that there are many sources that are more elementary than science.
Fourth and related, the study of scientism shows that the following kind of reasoning, that was once rather popular in neopositivist circles for reasons having to do with the infamous ‘verification criterion’, is often wrong: ‘If natural science doesn't tell that p, then p should not be believed, and should not be claimed to be known’. The natural sciences don't talk about wars, about what motivates people to act in certain ways, about what it is rational for humans to do, about what has value and what has not, about the meanings of words and sentences, etc. Yet we do seem to know quite a bit about these subject matters. From this it follows that in these cases the kind of reasoning under scrutiny is wrong.
Fifth, we can learn from the study of scientism that disciplinary imperialism should be resisted, lest we wind up with a view of ourselves and of the world that is radically at odds with the way we experience ourselves and the world. If certain approaches and methods work well in physics, there is no guarantee that they will work elsewhere as well, say in finance, or philosophy or art history. There is no guarantee that if some approach works well in biology, it must also work well in ethics. This should be shown case by case. The blind impulse to disciplinary imperialism should be resisted.
Footnotes
Disclosure statement
No potential conflict of interest was reported by the author(s).