1. Introduction Few people living in Western societies today would deny that science has great value. It is also widely believed, though, that the scope and value of science can be exaggerated; science has its boundaries and those boundaries should not be crossed (below, I say more about how this is to be understood). Philosophers or scientists who do cross these boundaries – or at least what many consider to be boundaries – are often referred to as subscribing to or practicing scientism. Here is a quote from William Provine that many, including some who defend scientism, would consider as an expression of adherence to scientism:
Modern science directly implies that the world is organized strictly in accordance with mechanistic principles. There are no purposive principles whatsoever in nature. There are no gods and no designing forces that are rationally detectable. (…) modern science directly implies that there are no inherent moral or ethical laws, no absolute guiding principles for human society. (…) human beings are marvelously complex machines. (…) when we die, we die and that is the end of us. (…) Free will as it is traditionally conceived – the freedom to make uncoerced and unpredictable choices among alternative possible courses of action – simply does not exist. (…) There is no ultimate meaning for humans.1 As I said, many would describe this as a clear case of scientism. But what precisely is scientism? If we consider the passage just quoted, we discover a wide variety of different claims: science implies that the world is purely mechanistic, that free will, as traditionally conceived, is an illusion, that there is no ultimate meaning for us, and so on. This gives rise to many questions. Is each of these theses an instance of scientism? Are there other kinds of scientism that are not found in this quote? Is there an underlying basic idea in virtue of which these claims are widely considered to be instances of scientism? How do different kinds of scientism relate to each other?
The aim of this paper is to provide a framework for answering questions like these by construing a conceptual map of scientism. By ‘scientism’ I mean, roughly, the view that the boundaries of the natural sciences2 should be expanded in order to encompass other academic disciplines and/or other realms of reality (e.g., our cognition, everything that exists, morality, or religion). What such expansion amounts to depends on the variety of scientism in question. It can mean, for instance, that only science can tell us what exists or science should completely replace common sense in a domain like morality.
To get sharper into focus what I mean by ‘scientism’, let me formulate three constrains on something to count as an instance of scientism. These constraints are based on how words such as ‘scientism’ and ‘scientistic’ are widely used in the literature. First, I treat scientism as a particular claim or thesis. This is not the only way one could think of scientism. One might also think of scientism as some kind of attitude, affection, stance, or still something else.3
For two reasons, I nonetheless prefer to treat scientism as a thesis. First, as evidenced by the quotations and references I give in this paper, scientism as a thesis is frequently found in the writings of scientists and philosophers. Second, it seems that every attitude, affection, or stance, at least if it is it to be up to discussion, can be translated into a thesis, namely the thesis that it is good to have that affection, attitude, or stance. No matter how one understands ‘scientism’, then, it will always imply some scientistic thesis or other.
Second, every instance of scientism puts the natural sciences or even a specific natural science, such as physics, centre stage. Every instance of scientism, then, is a claim about the relation that should obtain between the natural sciences or one particular natural science on the one hand and something else – another academic discipline or another realm of reality – on the other. This means that the claim that scientists themselves are somehow superior to other people4 falls outside the scope of this paper, even though this claim might in some way be related to scientism.
Third, even though the word ‘scientism’ is often used pejoratively, it need not be. For instance, James Ladyman, Don Ross, and David Spurrett, in their book Every Thing Must Go say expressis verbis that they adhere to scientism and go on to defend it in detail.5 Thus, to say that something is an instance of scientism is not thereby to say take a positive or negative stance towards the relevant assertion.6 In fact, I think that most philosophers and scientists will embrace at least some of the weaker versions of scientism described in this paper. Thus, a third constraint on something to count as an instance of scientism is that the thesis is formulated in such a way that it is up to discussion whether it is true.
When I say that I aim to provide a conceptual map of scientism, I mean that I aim to analyse the varieties in which scientism comes and how these varieties relate to each other, in order consequently to display the results of these analyses in a diagram. In doing so, I contrast my view with that of others who have written on scientism, especially Mikael Stenmark. It is not my aim to draw a map of all possible instances of scientism. Rather, I aim to draw a map of the most important varieties of scientism that we find in the literature. In construing the map I will use the words ‘variety’, ‘version’, and ‘instance’ of scientism. By a ‘variety of scientism’ I mean a species of the genus scientism: the variety entails scientism, but not vice versa. By a ‘version of scientism’ I mean a particular way of understanding a variety of scientism. And by an ‘instance of scientism’ I mean a particular person’s written or spoken verbal expression of her scientism.
The project of providing a conceptual map of scientism is important for at least two reasons. First, the word ‘scientism’ is often used in science, philosophy, and in the wider culture and frequently in a pejorative sense. However, it is often unclear what is meant when someone is labelled as an adherent of scientism. We can judge whether such labelling is correct only if we have some grip on the term ‘scientism’ and the varieties in which it comes. Second, it is important to see what an adherent of a particular variety of scientism is committed to. If a particular variety of scientism commits one to another variety with unpalatable implications, the position might be less plausible than initially thought. Or an adherent of scientism might falsely assume that a particular kind of scientism commits her to another kind of scientism. Whether this indeed the case is something that the conceptual map should make clear.
This paper is structured as follows. First, I sketch what I consider to be the main varieties of scientism. I distinguish between academic and universal scientism. Academic scientism comes in two varieties: methodological and reductive scientism, whereas universal scientism comes in four varieties: epistemological, ontological, moral, and existential scientism. (§ 2) Subsequently, I defend an account of how these varieties relate to each other. (§ 3) Also, I argue that there is a non-trivial set of necessary and sufficient conditions that a claim should meet in order to count as an instance of scientism. (§ 4) Finally, I draw the threads of this paper together by providing two figures that jointly constitute a conceptual map of scientism that displays the varieties of scientism and their interrelations. (§ 5)
2. Varieties of Scientism 2.1. Academic Scientism 1 and 2: Methodological and Reductive Scientism
The main distinction that we need to make is that between what I call academic scientism and universal scientism. Academic scientism is restricted to the academic disciplines,7 whereas universal scientism is meant to apply both inside and outside of the academy. We will see in a moment what these claims amount to.
The first distinction we need to make with respect to academic scientism is between methodological and reductive scientism. Whereas the methodological variety grants that, say, philosophy and psychology are proper academic disciplines that ask sensible questions, it asserts that they are so only if they adopt the methods of the natural sciences, such as observation and experimentation. Thus, the traditional questions of, say, theology or philosophy, can be answered only by using the methods of the natural sciences.8 The reductive version is stronger in that it claims that academic disciplines other than the natural sciences, such as the humanities, have nothing to add to the natural sciences if properly carried out. The questions asked in, say, psychology and philosophy, are nonsensical or obscure. We should abandon the subject matters of these disciplines altogether. For example, Otto Neurath gives a rather rhetorical statement of his view that that branch of philosophy which is called metaphysics should be completely reduced to physics, when he says:
how does the elimination of metaphysics proceed in practice? Men are induced to give up senseless sentences and freed from metaphysics. But must this always remain so? Must everyone in turn go through metaphysics as through a childhood disease – perhaps the earlier he gets it, the less dangerous it is – to be led back to unified science? No. Every child can in principle learn to apply the language of physicalism correctly from the outset, first in a crude form, then in a more refined and precise way.9 Neurath is rather explicit about his reductive academic scientism. Others, such as Patricia Churchland and Stephen Stich are less explicit. They argue that, since no consensus is forthcoming after two thousand years of discussion, we should abandon traditional philosophical problems, such as how knowledge is to be analyzed.10 These problems should be left aside altogether, since they cannot be solved by means of the natural sciences. The assumption here is, clearly, that only the natural sciences deliver what we are looking for (consensus) and that we should, therefore, give up any academic disciplines that do not employ the methods of those sciences.11
A second distinction that is relevant here is that between partial and full academic scientism. Whereas partial academic scientism makes a scientistic claim about only some of the academic disciplines that are distinct from the natural sciences, full academic scientism is a claim about all other academic disciplines than the natural sciences. Thus, Churchland’s and Stich’s claim that traditional epistemology should be replaced with neuroscience and cognitive psychology is an instance of partial scientism, whereas E.O. Wilson adopts a version of full scientism when he says: “It may not be too much to say that sociology and the other social sciences, as well as the humanities, are the last branches of biology waiting to be included in the Modern Synthesis [that is, neo-Darwinism; RP].”12 His idea seems to be that all academic disciplines should be reduced to the natural sciences, especially to biology.
A final distinction that we should make is that between weak and strong scientism. Whereas weak academic scientism claims that something should be reduced to the natural sciences in general, strong academic scientism claims that something should be reduced to one particular natural science. The main candidate disciplines for strong academic scientism seem to be physics and chemistry.13 Alex Rosenberg, for instance, says: “The phenomenal accuracy of its prediction, the unimaginable power of its technological application, and the breathtaking extent and detail of its explanations are powerful reasons to believe that physics is the whole truth about reality.”14 2.2. Universal Scientism 1: Epistemological Scientism
Let us now turn to universal scientism. I use the expression ‘universal scientism’ as a term of art in that it does not refer to a variety of scientism that is supposed to apply both within and outside of the academy. Some philosophers call this variety of scientism ‘academic-external scientism’. That seems misguiding, though. For, as we shall see, universal scientism applies not only to matters external to the university, but also within the academic realm. However, since universal scientism is as such not a claim about academic methodology or the reduction of one academic discipline to another, academic scientism is not a variety of universal scientism. In what follows, I distinguish four varieties of universal scientism.15
The first kind of universal scientism that we ought to distinguish is epistemological scientism. Epistemological scientism is scientism about the cognitive realm, both within and outside of the academy. There are numerous versions of epistemological scientism. Here are some of them:
All genuine knowledge is to be found only through (methods of) the natural sciences.16
The natural sciences provide the only reliable path to knowledge.17
All questions can in principle be answered by the natural sciences.18
Everything that can be known can be known through the natural sciences.19
There is a lot to be said about each of these theses. Here, I will only note that they are distinct theses. For instance, to claim that all genuine knowledge is to be found through the natural sciences is different from claiming that the natural sciences are the only reliable path to knowledge, for one might think that other methods than the natural sciences incidentally (unreliably) lead to knowledge.20 For practical reasons, in what follows, when I mention ‘epistemological scientism’ I confine myself to (a), the claim that all genuine knowledge is to be found only through the (methods of) natural sciences.
As with academic scientism, we can distinguish between full and partial epistemological universal scientism. Since this might sound somewhat paradoxical – does ‘universal’ not exclude ‘partial’? – let me explain this. I have used the word ‘universal’ in the rather restricted sense of ‘applying to both the academic and non-academic realms’. Now, one might think, for instance, that all knowledge about anything whatsoever is to be acquired by the natural sciences. That would count as full epistemological universal scientism. But one might also make the more restricted claim that all knowledge about, say, consciousness is to be acquired by the natural sciences. That would be a version of partial epistemological universal scientism. An example of the latter is what Bertrand Russell says about God and immortality:
God and immortality, the central dogmas of the Christian religion, find no support in science. (...) No doubt people will continue to entertain these beliefs, because they are pleasant, just as it is pleasant to think ourselves virtuous and our enemies wicked. But for my part I cannot see any ground for either. (…) no one of these hypotheses is more probable than any other; they lie outside the region of even probable knowledge, and therefore there is no reason to consider any of them.21 Russell’s point here seems to be that since the natural sciences cannot deliver any knowledge about God and immortality, surely anything else than the natural sciences will not deliver such knowledge either. This is a variety of partial epistemological universal scientism, because it is restricted to the supernatural realm (God and immortality). It does not say that the only knowledge we could possibly have about anything whatsoever is to be produced by the natural sciences. It is nonetheless universal because it applies both within and outside of the academy: if Russell is right, then theology and philosophy will not be able to deliver any knowledge about God or immortality either.
Finally, we can, of course, distinguish between weak and strong epistemological scientism. Whereas weak epistemological scientism says, for instance, that only the natural sciences deliver genuine knowledge, strong epistemological scientism claims that only, say, chemistry delivers knowledge. This distinction also applies to the other three varieties of universal scientism that I distinguish below, for reasons of brevity I will not repeat this point.
2.3. Universal Scientism 2: Ontological Scientism
The second variety of universal scientism is not a claim about our knowledge, rational beliefs, or true beliefs, but about what does and does not exist. Now, some philosophers, such as Roger Trigg, have taken this variety of scientism to amount to the claim that only those things exist which are at some point discovered by science.22 As I said in the Introduction, though, the aim of my conceptual map is to provide an overview of options that could plausibly be defended, not an overview of any possible position. And the claim that only those things exist that are discovered by science at some point is clearly implausible, given that there are vast stretches of the universe that we might never be able to explore.
This is not to deny that there have been scientists and philosophers who seem to claim something as strong as the kind of scientism described by Trigg. According to Wilfrid Sellars, for instance, “in the dimension of describing and explaining the world, science is the measure of all things; of what is that it is and of what is not that it is not.”23 As it seems to me, though, the most plausible way to interpret this sentence is not as saying that only those things exist that are acknowledged by science, but as saying that only if something is acknowledged by science we should believe or we can know that it exists. Thus, this is really an instance of epistemological scientism. I think this is not at all unusual, even though adherents of scientism sometimes conflate epistemological and ontological scientism.24
Can we nonetheless distinguish ontological scientism as a separate variety of scientism? I think we can. Take the opening statement of Carl Sagan’s classic book Cosmos: “The cosmos is all that is or ever was or ever will be.”25 Although he does not say this explicitly, the idea is clearly that every aspect of the cosmos can in principle be investigated by science. Other things, such as free will, universals, and immaterial souls do not exist, for all that exists is matter and can be investigated by science. Full versions of ontological scientism will be rare, but partial varieties will not. On partial ontological scientism, a specific kind of thing is nothing but (a collection of) those things acknowledged by the natural sciences. Undoubtedly, the most popular variety of partial ontological scientism is scientism about human beings. Carl Sagan, for instance, describes himself as a collection of water, calcium and organic molecules.26
Francis Crick has called the idea, which he himself advocates, that we are nothing but a pack of neurons ‘The Astonishing Hypothesis’. It is astonishing, because it means that our memories, ambitions, beliefs, desires, choices, sorrows, and so on are nothing but the collection or behaviour of a large sum of nerve cells and their associate molecules.27 Another version of partial ontological scientism is the idea that free will is an illusion because science can explain every decision without appeal to free will, as advocated, for instance, by William Provine in that quote that I gave in the Introduction.
2.4. Universal Scientism 3: Moral Scientism
A third variety of universal scientism is moral scientism. There are two different varieties of moral scientism. On the first variety, the natural sciences lead or will lead us to the good life. Here are some versions of it:
The natural sciences guide us towards the morally good life.28
Common sense morality should be replaced with scientific morality.29
Our moral personal and social problems can be solved by natural sciences.30
The second variety says something rather different. Here the basic idea is that science shows us or makes it sufficiently probable that morality is an illusion. Here are some versions of it:
Science shows us that morality is an illusion.
Science shows us that good and evil are merely social conventions.
Science shows us that moral intuitions and beliefs are nothing but evolutionarily adaptive features of humans.
The second variety of moral scientism is clearly a case of partial ontological scientism. I will nonetheless also treat it as a variety of moral scientism, since it is explicitly about morality. I will call these two varieties respectively the R-variety (from ‘replacement’) and I-variety (from ‘Illusion’).
According to some philosophers, such as Stenmark, there is a further variety of moral scientism, namely the claim that evolutionary theory can explain our moral sense. I agree that adherents of moral scientism usually also make this claim. But it seems to me misguided to take this claim itself to be an instance or part of moral scientism. Our moral beliefs are a natural phenomenon and it is, thus, not at all controversial that there might be some kind of scientific explanation for many or all of our moral beliefs, in terms of evolutionary theory or cultural history. In order for a view to count as moral scientism, it should make a stronger claim. According to E.O. Wilson, for instance, scientists and humanists should seriously consider removing ethics from the hands of philosophers, in order to biologize it.31 The idea seems to be that biological principles can be applied in the social realms and that they can be used to justify and not merely explain certain moral norms and values.32
Now, one might think that moral scientism is partial by its very nature, given that it is restricted to the moral realm. But this is mistaken. In the same way as there can be full and partial academic scientism, even though academic scientism is restricted to the academic realm, there can be full and partial moral scientism, even though moral scientism is restricted to the moral realm. One might think, for instance, that science can replace some of our morality, but not all of it, since we need to start from some moral intuitions. It does seem true, though, that on its I-variety, moral scientism does not come in full and partial varieties. It seems hard to defend that, say, moral obligations are illusions, but that moral intuitions can nonetheless be true. Hence, the R-moral scientism can be full or partial, whereas I-moral scientism can only be full scientism.
2.5. Universal Scientism 4: Existential Scientism
The final variety of universal scientism that I would like to distinguish is existential scientism. As with moral scientism, this variety of scientism comes in two rather different varieties. The first variety, which I call the R-variety, says that science should replace religion, mythology, secular ideologies, such as fascism and Marxism, and other non-scientific ways of answering our existential questions. Here are some versions of existential scientism:
Science should replace traditional religions and secular ideologies.33
Salvation can be achieved by (the methods of) science alone.34
Richard Dawkins articulates existential scientism when he says that the answers we give to the big questions of life are meaningless unless they are informed by natural science, especially evolutionary biology. The entire intellectual traditions of, say, ancient Greek philosophy and Medieval scholastics are without worth, because they are not based on scientific research.35
The second variety, which I will call the I-variety says that the idea that there is ultimate meaning or purpose in life is illusory.
Some versions of the R-variety of existential scientism are full, others partial. One might think that science can answer all our existential questions or that science can replace all aspects of traditional religions. But one might also think that science can only replace certain aspects of traditional religions and secular ideologies, such as their answers to questions about the ultimate origin of human beings, but not answers to questions about meaning and purpose in life. Even its I-variety seems to admit of full and partial varieties. One might think that science shows that God is an illusion, but that there is nonetheless objective meaning and purpose. One might think that properties concerning value and meaning supervene on natural properties, even though there is no God.
Some have included other theses under the umbrella of existential scientism. According to Stenmark, for instance, the idea that evolutionary theory can explain religious beliefs and the view that it can undermine religious belief also count as part of existential scientism. It is true that adherents of existential scientism are likely to adopt these theses as well, but we should not conclude from that that these theses are instances of existential scientism. Several religious scientists and philosophers embrace the thought that our religious beliefs are produced by a Hyperactive Agency Detection Device and many of them would agree that certain religious ideas, such as that the earth is 6,000 years old, have been undermined by science. Still, they do not seem thereby to count as adherents of scientism.
2.6. Further Varieties of Universal Scientism?
Are these four versions of universal scientism exhaustive? Let me discuss two proposals for further varieties of universal scientism.
First, Stenmark distinguishes the thesis that a belief is rational only if it is the deliverance of the natural sciences as a separate variety of scientism and calls it rational scientism. He interprets ‘epistemological scientism’ as the claim that only the natural sciences can deliver knowledge and rightly points out that that thesis is conceptually distinct from rational scientism. He fails to acknowledge, though, that the two are closely related to each other. Imagine that you believe that only the natural sciences can deliver knowledge. Imagine also that you consider some proposition p that you believe you have no scientific evidence for. Then, if you are rational, you will believe that you cannot know that p. You will realize that if you nonetheless believe that p, you might turn out to be lucky and hold a true belief, but you will also realize that in such a case you will not know that p. But if you are aware of the fact that you cannot know that p, then how could you possibly rationally believe that p? On most accounts of rationality, the belief that one cannot know that p provides a defeater for rationally believing that p. Someone who is sufficiently rational will realize that if (she believes that) she cannot know that p, then she cannot rationally believe that p. This means that epistemological scientism, in conjunction with some highly plausible principles about knowledge and rationality, entails rational scientism. And that means that we have good reason not to treat rational scientism as a separate variety of universal scientism, but as a version of epistemological scientism.
Second, Stenmark also distinguishes what he calls axiological scientism. The idea here is that the natural sciences are more valuable than other ways of learning. Those other ways usually include at least the humanities, but some adherents of scientism add politics, sports, art, literature, and philosophy to the list.36 Some philosophers seem to understand scientism almost exclusively along the lines of axiological scientism. According to Tom Sorell, for instance, scientism is the view that natural science is much the most valuable part of human learning.37
Is axiological scientism truly a separate variety of scientism? I doubt it is. For, one might ask in what sense the natural sciences are supposed to be more valuable than other ways of learning. In other words, what kind of value is referred to in axiological scientism? Three possibilities come to mind: such value could be epistemic, moral, or existential. In other words, the natural sciences can be more valuable in that they are more likely to lead to knowledge, in that they guide us in leading the good life, or in that they help us to meet our existential needs. But, clearly, axiological scientism will then be reducible to respectively epistemological, moral, and existential scientism. The only kind of value that I can think of that is not reducible to epistemological, moral, or existential value is aesthetic value; the value beautiful music, painting, and architecture have. But it seems that (virtually) no one has made the rather bold assertion that science is aesthetically more valuable than other ways of learning. That means that we should not treat axiological scientism as an additional variety of scientism.