I apologise to those who find this kind of thing boring, but I simply can’t let this go, so here I am with scientism once again, and I’ve spent too long puzzling over this just throw it all away. The length, I’m afraid, is heading into the stratosphere, even for me! Anyway, I promised Jerry that I would come back to it.
I have been reading Jason Rosenhouse’s posts on scientism over at Evolution Blog, here and here. I don’t want to make heavy weather of this, but, again, I think the point is being missed. Jason also has another post about scientism, in response to Kitcher’s New Republic essay “The Trouble with Scientism” (which you can access here). And, of course, there are Jerry Coyne’s comments on Kitcher’s essay, which you can access here and here. (And, while I do not link to them, Massimo Pigliucci has a number of posts on scientism over at Rationally Speaking.) So, it’s not as though the issue has not raised a number of concerns. I pick Rosenhouse and Coyne because I admire them, and because I owe both of them so much. (Along with Ophelia Benson, they have been my mentors — unknown, perhaps, to them — on how to go about the business of blogging.) I do think, though, that on the subject of scientism they seem to me to miss the point.
The general response to Kitcher’s concerns about scientism are to say, basically, that those disciplines that do produce knowledge, produce knowledge because they use the same methods, the same “ways of knowing” as science. The implication is that, where a discipline, like archaeology, ethnography, economics, history, etc. is successful in advancing our knowledge, it does so by invoking the same methods, and obtaining the same kinds of evidence, as does science. Here is how Jerry puts it in “The trouble with ‘The trouble with scientism’”:
If you look at [Kitcher's] examples of where scholars have produced increasing understanding and progress, it is in disciplines like history, economics, ethnography, and archaeology — fields that rely on the same “ways of knowing” as does science.
We’ll come back to that locution “ways of knowing” in a moment, but at this point I want to quote a similar opinion from Jason Rosenhouse (from his second post on the silliness of scientism):
The humanities are under attack, but not because of anything scientists are doing. It certainly has nothing to do with anything Philip Kitcher discussed in his article for The New Republic. (That was the one where he made the strange argument that scientism is false because the humanities are more like sciences than is sometimes acknowledged, as we have discussed.) [my italics]
Now, it seems to me, based on these two remarks, that we need to go take a closer look at the idea of ways of knowing.
The key question is: What is meant by the locution “ways of knowing”? It is important to notice, while discussing this question, that the question itself is designed to rule out certain things as capable of being known, things for which there are, in fact, no ways of knowing at all. Religious or theological knowledge is what is chiefly in mind. The fundamental thought here is that if you construe the idea of “scientific ways of knowing” broadly enough, you’ll capture everything that is capable of being known, and exclude everything for which claims to knowledge have been made, but which, on examination, turn out to be bogus claims, religious claims to knowledge in particular.
The clearest expression of this is in Jason Rosenhouse’s first post on the silliness of scientism, where he say this:
It seemed obvious to me then, and still seems obvious to me now, that the only reliable way of obtaining knowledge about the physical world is by applying the common sense investigative techniques that everyone applies in their everyday lives. You gather the facts, formulate theories, test your theories by acquiring more facts, and so on. Since these methods obtain their most precise and fruitful applications in the work of scientists, it seems reasonable to say that science is the only way of knowing.
However, there’s a fallacy here. Consider the argument:
- Knowledge of the physical world is only reliably obtained by the commonsense investigative techniques that everyone uses in their daily lives.
- These techniques consist in observing facts, constructing theories to explain the facts, testing the theories by further observation of facts, and so on.
- Science uses these techniques most successfully in obtaining precise, progressively accumulating knowledge of the physical world.
- Therefore, science is the only way of knowing.
The premises do not support the conclusion. In order to support this conclusion we need a premise to the effect that there is no knowledge other than knowledge of the physical world.
Now, Jerry Coyne thinks he has this covered by claiming, as Jason Rosenhouse does, that we need to construe ‘science’ broadly — as he puts it in “The trouble with ‘The trouble with scientism’” — “as the use of reason, empirical observation, doubt, and testing as a way of acquiring knowledge,” and then adding to this the claim that “ultimately all phenomena, including human social interactions, come down to the motions of molecules.” He grants that “we’ll never be able to understand our society using the tools of physics alone,” because
[t]hings as complicated as human society (nay, any society) are higher-order emergent properties that often demand their own methods. [last italics mine]
It’s not altogether clear what is meant here by “their own methods,” but it is hard to escape the suspicion that what we have here are other ways of knowing (with their own methods). Recall Rosenhouse’s claim that
the only reliable way of obtaining knowledge about the physical world is by applying the common sense investigative techniques that everyone applies in their everyday lives.
But if what we are dealing with has emergent properties, so that, while it is consistent with the laws of physics, it will never be able to be understood by means of physics alone, then it seems that something other than “knowledge of the physical world” would have to be invoked here, and it would, perhaps, use different methods than those used by science.
It is precisely this kind of complexity that Kitcher speaks about in his New Republic article. Where things are incredibly complex, even in the physical world, as, for example, when we are dealing with climate, or the earth sciences:
The enterprises that we lump together are remarkably various in their methods, and also in the extent of their successes. The achievements of molecular engineering or of measurements derived from quantum theory do not hold across all of biology, or chemistry, or even physics. Geophysicists struggle to arrive at precise predictions of the risks of earthquakes in particular localities and regions. The difficulties of intervention and prediction are even more vivid in the case of contemporary climate science.
And then, of course, we have human beings themselves, their societies, cultures, literatures, arts, histories, and so on, which are so bewilderingly complex that there are points at which it makes sense to say that we are dealing with things which are not knowable as parts of the physical world, as such, because they have to do with meaning, purposes, hopes, fears, expectations, and so on, and not with empirical observations. They are parts of the human world, or the social world. (Perhaps we need something like Popper’s three worlds distinction.)
The central problem here is that Coyne and Rosenhouse cannot really say what they want to say without breaking their own rules. Knowledge, they claim, is scientific, but in saying that knowledge is scientific they are claiming to know something which is not scientific, namely, the claim that knowledge itself is scientific. In his discussion with Julian Baggini, Lawrence Krauss says:
Ultimately, I think our understanding of neurobiology and evolutionary biology and psychology will reduce our understanding of morality to some well-defined biological constructs.
“Yep, that’s pretty bad,” says Rosenhouse, and then he goes on in more detail to say:
I’m not sure why the notion that morality must be entirely a scientific question has become a thing among people like Krauss and Sam Harris. Surely it’s just obvious that you need something more than scientific facts alone to resolve moral questions.
And I agree. But when Rosenhouse goes on to say –
But here’s the thing. Usually when someone makes a statement that you think is mistaken, the proper response is, “He’s wrong! Here’s why …” You don’t normally accuse him of being in thrall to an ism. That’s a totally different level of criticism.
– again I have to agree. For, it really is a totally different level of criticism, and it’s not just wrong. The problem is precisely that Krauss and Rosenhouse want to say that science is the only way of knowing. Whatever morality or history (except in so far as history deals with physical evidence) may be, they don’t deal in knowledge. We can’t say, after we’ve finished making a moral argument, that we know it is wrong to torture babies for fun, because that’s not a scientific statement and cannot be confirmed by the methods of science. The problem is that Rosenhouse cannot reasonably both say, as he does, that “it seems reasonable to say that science is the only way of knowing,” and then say that he doesn’t understand why for people like Krauss morality’s being entirely scientific has become such an issue with some people; because if science is the only way of knowing, then there simply is no moral knowledge, in which case Krauss’s saying that morality might some day be reduced to biological constructs would not be pretty bad at all.
The other problem here is that empirical evidence, as Kitcher points out in his piece on “The Trouble with Scientism,” cannot stand on its own. It only makes sense within a theoretical context. For instance, the identification of a particle — say, the Higgs Boson – in the Large Hadron Collider in Switzerland, only makes sense within an incredibly complex theory that most people do not and cannot understand (except at several removes by way of a popularisation of the theory). As Kitcher says:
Successful sciences are collections of models of different types of phenomena within their domains. The lucky ones can generate models that meet three desiderata: they are general, they are precise, they are accurate. [my italics]
And he goes on to point out that not all natural sciences are lucky in this respect. And the models themselves, in terms of which an observation is counted as evidence, are not themselves observations, and do not themselves constitute empirical data. So what is known in science is the theory, as Richard Dawkins often points out. The truth of the theory is based on its confirmation by data — which is when, to use Dawkins’ coinage, it becomes a theorum. But no one ever observed the theory. This in itself is not observed, and, on some occasions in the history of science, theories believed to be based firmly on physical evidence have turned out to be false, as Kitcher points out:
No less a figure than Maxwell even characterized the ether as “the best confirmed entity in natural philosophy” (by which he, like his contemporaries, meant “natural science”).
The point is that scientific theories themselves are cultural products, and as firmly based as they may be in the empirical data, they are still at one remove (inductively) from the data themselves. They are structures of meaning in terms of which, in the localised regions of the empirical world to which they apply, we understand the natural world.
In this respect history is like science, for it must deal with facts just as science does, but it takes note of facts in very different ways, as Kitcher tries to point out. For instance, speaking of the social and psychological sciences, Kitcher says (and here we must quote at length):
Human social behavior arises, in a complex social context, from the psychological dispositions of individuals. Those psychological dispositions are themselves shaped not only by underlying genotypes, but also by the social and cultural environments in which people develop. Cultural transmission occurs in many animal species, but never to the extent or to the degree to which it is found in Homo sapiens. Human culture, moreover, is not obviously reducible to a complex system of processes in which single individuals affect others. Rigorous mathematical studies of gene-cultural coevolution reveal that when natural selection combines with cultural transmission, the outcomes reached may differ from those that would have been produced by natural selection acting alone, and that the cultural processes involved can be sustained under natural selection. Whether this happens in a wide variety of areas of human culture and domains or is relatively rare is something nobody can yet determine. But culture appears to be at some level autonomous and in some sense irreducible, and this is what scientism cannot grasp.
This is just where Jason Rosenhouse’s claim about Kitcher’s article, that it includes “the strange argument that scientism is false because the humanities are more like sciences than is sometimes acknowledged,” breaks down. What Kitcher is saying here is that there seems to be a level at which ”human inquiry needs a synthesis, in which history and anthropology and literature and art play their parts,” in which, therefore, we cannot understand (that is, we cannot know) about some aspects of being human without including contributions from these very different humanistic studies, and are, in this way, not like sciences at all.
In other words, Kitcher is suggesting that there are different kinds of knowledge or ways of knowing that are appropriate to different domains. Mathematics and logic, for example, are domains of knowledge in many respects quite unlike science. And, just as we can know how to ride a bicycle, what it feels like to ride on a roller-coaster, swim in the ocean, meet face to face with a shark, etc., history can help us know what it would feel like to be a Roman citizen at the time that the empire fell. The latter experience, as Tom Holland points out, was almost seismic in its effects. In Holland’s words:
“The greatest, perhaps, and most awful scene, in the history of mankind”: so Gibbon described his theme. He was hardly exaggerating: the decline and fall of the Roman empire was a convulsion so momentous that even today its influence on stories with an abiding popular purchase remains greater, perhaps, than that of any other episode in history.
The point here is that we are dealing with a completely different dimension of human knowledge. Dependent though it may be on ascertainable facts recorded in documents or artefacts, history and anthropology are places, as Kitcher says, where new concepts are forged, and large social changes take place — which is why, incidentally, that people object to bringing all knowledge under the heading of science. We are dealing with knowledge of the social world or the human world, where feelings and beliefs and values and fears all come into play. Indeed, says Kitcher:
Once the intertwining of human inquiry with social change has been recognized, it is easy to see why history and ethnography demand constant rewriting.
For we have here entered the realm of cultural interpretation. We could not write history as Gibbon did, because, as Kitcher says,
his words are not ours — it would be odd to speak as he does of the “licentiousness,” “prostitutions,” and “chastity” of the empress Theodora. If our questions are different, it is because we live in a very different culture, one that his history helped to bring about.
We interpret the facts differently, because we see and experience things differently. Things have a different valence, a different resonance. The same facts can produce quite different historical interpretations. Facts by themselves do not constitute history. In order to explain the facts — this is where theory and interpretation come in — they need to be put into an explanatory, narrative context.
Consider the example with which Kitcher begins his article, where he speaks of the fire-bombing of Dresden towards the end of the war in Europe in February 1945:
Since the 1960s, historians have worked — and debated — to bring into focus the events of the night of February 13, 1945, in which an Allied bombing attack devastated the strategically irrelevant city of Dresden. An increased understanding of the decisions that led to the firebombing, and of the composition of the Dresden population that suffered the consequences, have altered subsequent judgments about the conduct of war. The critical light of history has been reflected in the contributions of novelists and critics, and of theorists of human rights.
He returns to Dresden later in the article, where he says this:
Although studies such as these make no pretense at generality, their impact can be very large. They can unsettle the categories that are taken for granted in all kinds of decisions, from mundane reflections about how to respond to other people to large matters of social policy. “Collateral damage,” for example, comes to seem an inappropriate way to talk about the victims of the Dresden fire-bombing.
History is rewritten, and with it, our concept of war changes. We come to understand, and therefore to know, things in ways undreamt of by our predecessors. I want to do this with assisted dying, to show that we do not need to regard assisted dying in terms of the old moral category of suicide, for example. It can be placed within a new social context, in which autonomy and individuality play a larger role, and in which personal responsibility for major life decisions is emphasised. Suggesting that the methods of history are the same as the methods of science is a misrepresentation of what historians do. Cultural context and meaning transform the significance of the facts. While their studies are critical, and depend on facts that can be verified, the element of interpretation which is introduced, as Richard Evans says, in his book, In Defence of History, “where the facts are converted into evidence (that is, facts used in support of an argument),” (76; my italics) is vital to our understanding of what historians do, and how they reach the conclusions they do — conclusions which are not, Kitcher says, independent on the historical-cultural context in which the historian works.
Of course, in science also, empirical observations must be placed in an interpretive context in order to be understood. This interpretive context is known as theory. Before he had the theory, Darwin’s observations may have constituted knowledge of facts, but not scientific knowledge. Scientific theory, unlike historical, anthropological or literary theory, is largely independent of cultural change. As Jerry Coyne says in a brand new post:
I don’t give a hoot whether the USA beats all other nations in the quality and output of its scientists. That, to me, is a form of chauvinism, and science, being an international venture, should be promoted everywhere. A rising tide lifts all boats. We should try to eliminate gender bias not because it will make the U.S. more competitive, but simply because it’s the right thing to do. [my italics]
Even though history and anthropology are less international in scope, they do provide knowledge, a different kind of knowledge than that provided by science, less assured, more dependent on the local interpretive context (culture). The same goes for our moral knowledge; but here at the end we are still left with the question of how Jerry knew (on his terms)that eliminating gender bias is the right thing to do?