Errol Morris, Kuhn & the Ashtray
Errol Morris published the fourth installment of his five-part personal essay on Kuhn in today’s NY Times. All of it is worth reading (except perhaps part III, which drags on somewhat), and I definitely encourage everyone to take a look!
What is Morris’ beef with Kuhn? In a nutshell, he thinks that Kuhn provided volatile ammunition for “postmodernists” (that’s Morris’ term): people who want to deny there is a truth (rather than many truths), a real world (rather than multiple realities), and a compelling distinction between ethical and unethical actions (rather than just social mores). Now, I suspect that Morris is mostly tilting at windmills here, or, at the very least, that his argument is about ten years behind the times. But for historians of science, I think, it is worth thinking about what’s got him so upset.
Morris’ central objection centers on the relationship between paradigms and incommensurability. What is a paradigm and what does it mean for two paradigms to be incommensurable? In The Structure of Scientific Revolutions, Kuhn equivocates on both scores. But as far as I can tell, paradigms are basically how we understand the world. They are general frameworks we use to organize our experience. For example, Kuhn invokes Einstein’s theory of general relativity. Initially, many people rejected the idea that space has a curvature. This is because in the existing paradigm, space is simply not the kind of thing that can be curved; to say that space is curved is like saying triangles are virtuous. This is where incommensurability comes in: if two paradigms are incommensurable then the conceptual content of one cannot be articulated in terms of the other (“space has a curvature” simply makes no sense on the old interpretation of “space”). Hence, people operating in different paradigms are going to have a hard time talking sense to each other. Sometimes, Kuhn went so far as to say that people who operate in different paradigms literally live in different worlds!
Morris dislikes this line of thinking because it suggests the search for truth is bound to be forever elusive. Differently put, the history of science is just one damn thing after another! If there is no way to get outside of a paradigm, then there can be no basis for comparison between different paradigms. Hence, we cannot say that one is better than another. Paradigm shifts are a social, political, or perhaps psychological process. They are not rational. At the end of his latest installment, Morris says that in this, Kuhn sided with the later Wittgenstein: “we could agree that the earth is flat and that would make it so.” Morris goes on to quote from Structure itself: “We may … have to relinquish the notion … that changes of paradigm bring scientists … closer and closer to the truth.”
Earlier, I said that Morris was tilting at windmills. Let me say something more about what I mean by this. Historians of science often like to remark on the irony that although Kuhn probably did more damage to the logical empiricist research tradition than anyone, Structure was initially published as part of the International Encyclopedia of Unified Science, which also counted Carnap, Hempel, Nagel, and Neurath among its contributors. How are we to make sense of this fact? There’s a couple of things we can say. First, it was simply the case that logical empiricists were big-shot philosophers at the time, so even if you disagreed with them it made sense to submit your work to their publication venues. This may indeed be what happened. Still, although Kuhn certainly disagree with much of the positivist tradition he did share a deep set of assumptions with them. Perhaps the most important of these is a focus on linguistic structures. Both Kuhn and the logical empiricists had a deep conviction that language provides the interface between the world and ourselves. Hence, if you want to understand everything from metaphysics to the philosophy of science the thing to do is to focus on language.
Let’s just look at Rudolf Carnap as an example. In 1950 he published a fascinating paper in the Revue International de Philosophy called “Empiricism, Semantics, and Ontology.” I don’t want to get bogged down by the details of his argument, so let’s focus on one idea in particular: that of a linguistic framework. Carnap introduced this concept because he wanted to distinguish between two kinds of questions. The first kind ask things such as “do unicorns exist?” or “is there a prime number above 100?”. The second are more general, and inquire whether there are physical objects or abstract entities. He designated the former internal questions and the latter external questions. The notion of a linguistic framework was introduced to distinguish between them. He wrote: “If someone wishes to speak in his language about a new kind of entities, he has to introduce a system of new ways of speaking, subject to new rules; we shall call this procedure the construction of a linguistic framework for the new entities in question.”
Carnap’s move here is a very attractive one. All he is trying to do is make sense of what is required for people to be able to talk meaningfully. For example, asking questions about prime numbers requires that we have some framework with which to talk about numbers. In the case of arithmetic, this might be predicate logic enriched with Peano axioms. These give you the rules for how to adjudicate a question like “is there a prime number greater than 100?”. Or, if you want to ask about the existence of some particular physical object we turn to the framework of natural science. The point is, to quote from Carnap again, that to “recognize something as a real thing or event means to succeed in incorporating it into the system of things … according to the rules of the framework.”
So much for internal questions. What about external ones like “do numbers exist?”. These, Carnap thinks, are questions that cannot be answered within any given framework. Rather, they are questions about the framework itself. Now, the whole point of frameworks is that they give you the rules with which to adjudicate ontological questions. Without the framework there are no rules and hence no way to answer the question. To quote from Carnap one last time, “To accept the thing world means nothing more than to accept a certain form of language, in other words, to accept rules for forming statements and for accepting or rejecting them.” Moreover, “the thesis of the reality of the thing world cannot be among these statements, because it cannot be formulated in the thing language or, it seems, in any other theoretical language.” The choice of frameworks, he thinks, is, if not exactly arbitrary then certainly heuristic: “The efficiency, fruitfulness, and simplicity of the use of the thing language may be among the decisive factors” in choosing to adopt it.
I don’t mean to suggest the Kuhn ripped off Carnap and simply re-christened frameworks as paradigms. But I do think there is a deep set of shared assumptions at play here. The most important one, the one that I want to highlight here, is that it is language or some other analogous, abstract, rule-governed structure that determines our experience of the world. We are, in some sense or another, trapped within language, whether we want to call it a paradigm or a framework. I think this is what Kuhn might have meant with his infamous remark that two scientists operating in different paradigms do not share the same world: our language constitutes our world because it provides the rules according to which we accept or reject the existence of certain entities, events, and relations.
Let’s follow Morris one step further in the argument. I do think he is onto something when he makes the connection to postmodernism. This is perhaps even more ironic, because postmodernists (whoever they were) certainly would have been terribly unhappy to be mentioned in the same breath as Rudolf Carnap. But I do think the two schools of thought share some very deep assumptions about language and ontology, namely, the notion that our world is never simply given to us, but rather, that our experience of it is fundamentally structured by language or some other abstract symbolic system. Now, of course there are questions about how this system works—the post-structuralists, for example, would disagree that it is fundamentally logical or rule-governed—but the role of abstract systems is taken to be pervasive across the board. The radical move that postmodernism made is to sever all connections between the abstract system and what it represents, to emphasize the vast gulf that separates signifiers from their signifieds. Or, to put it another way, although the fundamental epistemology remained the same--a picture theory of truth--a change took place in that the picture was made more real than what it represents.
What is Morris’ beef with Kuhn? In a nutshell, he thinks that Kuhn provided volatile ammunition for “postmodernists” (that’s Morris’ term): people who want to deny there is a truth (rather than many truths), a real world (rather than multiple realities), and a compelling distinction between ethical and unethical actions (rather than just social mores). Now, I suspect that Morris is mostly tilting at windmills here, or, at the very least, that his argument is about ten years behind the times. But for historians of science, I think, it is worth thinking about what’s got him so upset.
Morris’ central objection centers on the relationship between paradigms and incommensurability. What is a paradigm and what does it mean for two paradigms to be incommensurable? In The Structure of Scientific Revolutions, Kuhn equivocates on both scores. But as far as I can tell, paradigms are basically how we understand the world. They are general frameworks we use to organize our experience. For example, Kuhn invokes Einstein’s theory of general relativity. Initially, many people rejected the idea that space has a curvature. This is because in the existing paradigm, space is simply not the kind of thing that can be curved; to say that space is curved is like saying triangles are virtuous. This is where incommensurability comes in: if two paradigms are incommensurable then the conceptual content of one cannot be articulated in terms of the other (“space has a curvature” simply makes no sense on the old interpretation of “space”). Hence, people operating in different paradigms are going to have a hard time talking sense to each other. Sometimes, Kuhn went so far as to say that people who operate in different paradigms literally live in different worlds!
Morris dislikes this line of thinking because it suggests the search for truth is bound to be forever elusive. Differently put, the history of science is just one damn thing after another! If there is no way to get outside of a paradigm, then there can be no basis for comparison between different paradigms. Hence, we cannot say that one is better than another. Paradigm shifts are a social, political, or perhaps psychological process. They are not rational. At the end of his latest installment, Morris says that in this, Kuhn sided with the later Wittgenstein: “we could agree that the earth is flat and that would make it so.” Morris goes on to quote from Structure itself: “We may … have to relinquish the notion … that changes of paradigm bring scientists … closer and closer to the truth.”
Earlier, I said that Morris was tilting at windmills. Let me say something more about what I mean by this. Historians of science often like to remark on the irony that although Kuhn probably did more damage to the logical empiricist research tradition than anyone, Structure was initially published as part of the International Encyclopedia of Unified Science, which also counted Carnap, Hempel, Nagel, and Neurath among its contributors. How are we to make sense of this fact? There’s a couple of things we can say. First, it was simply the case that logical empiricists were big-shot philosophers at the time, so even if you disagreed with them it made sense to submit your work to their publication venues. This may indeed be what happened. Still, although Kuhn certainly disagree with much of the positivist tradition he did share a deep set of assumptions with them. Perhaps the most important of these is a focus on linguistic structures. Both Kuhn and the logical empiricists had a deep conviction that language provides the interface between the world and ourselves. Hence, if you want to understand everything from metaphysics to the philosophy of science the thing to do is to focus on language.
Let’s just look at Rudolf Carnap as an example. In 1950 he published a fascinating paper in the Revue International de Philosophy called “Empiricism, Semantics, and Ontology.” I don’t want to get bogged down by the details of his argument, so let’s focus on one idea in particular: that of a linguistic framework. Carnap introduced this concept because he wanted to distinguish between two kinds of questions. The first kind ask things such as “do unicorns exist?” or “is there a prime number above 100?”. The second are more general, and inquire whether there are physical objects or abstract entities. He designated the former internal questions and the latter external questions. The notion of a linguistic framework was introduced to distinguish between them. He wrote: “If someone wishes to speak in his language about a new kind of entities, he has to introduce a system of new ways of speaking, subject to new rules; we shall call this procedure the construction of a linguistic framework for the new entities in question.”
Carnap’s move here is a very attractive one. All he is trying to do is make sense of what is required for people to be able to talk meaningfully. For example, asking questions about prime numbers requires that we have some framework with which to talk about numbers. In the case of arithmetic, this might be predicate logic enriched with Peano axioms. These give you the rules for how to adjudicate a question like “is there a prime number greater than 100?”. Or, if you want to ask about the existence of some particular physical object we turn to the framework of natural science. The point is, to quote from Carnap again, that to “recognize something as a real thing or event means to succeed in incorporating it into the system of things … according to the rules of the framework.”
So much for internal questions. What about external ones like “do numbers exist?”. These, Carnap thinks, are questions that cannot be answered within any given framework. Rather, they are questions about the framework itself. Now, the whole point of frameworks is that they give you the rules with which to adjudicate ontological questions. Without the framework there are no rules and hence no way to answer the question. To quote from Carnap one last time, “To accept the thing world means nothing more than to accept a certain form of language, in other words, to accept rules for forming statements and for accepting or rejecting them.” Moreover, “the thesis of the reality of the thing world cannot be among these statements, because it cannot be formulated in the thing language or, it seems, in any other theoretical language.” The choice of frameworks, he thinks, is, if not exactly arbitrary then certainly heuristic: “The efficiency, fruitfulness, and simplicity of the use of the thing language may be among the decisive factors” in choosing to adopt it.
I don’t mean to suggest the Kuhn ripped off Carnap and simply re-christened frameworks as paradigms. But I do think there is a deep set of shared assumptions at play here. The most important one, the one that I want to highlight here, is that it is language or some other analogous, abstract, rule-governed structure that determines our experience of the world. We are, in some sense or another, trapped within language, whether we want to call it a paradigm or a framework. I think this is what Kuhn might have meant with his infamous remark that two scientists operating in different paradigms do not share the same world: our language constitutes our world because it provides the rules according to which we accept or reject the existence of certain entities, events, and relations.
Let’s follow Morris one step further in the argument. I do think he is onto something when he makes the connection to postmodernism. This is perhaps even more ironic, because postmodernists (whoever they were) certainly would have been terribly unhappy to be mentioned in the same breath as Rudolf Carnap. But I do think the two schools of thought share some very deep assumptions about language and ontology, namely, the notion that our world is never simply given to us, but rather, that our experience of it is fundamentally structured by language or some other abstract symbolic system. Now, of course there are questions about how this system works—the post-structuralists, for example, would disagree that it is fundamentally logical or rule-governed—but the role of abstract systems is taken to be pervasive across the board. The radical move that postmodernism made is to sever all connections between the abstract system and what it represents, to emphasize the vast gulf that separates signifiers from their signifieds. Or, to put it another way, although the fundamental epistemology remained the same--a picture theory of truth--a change took place in that the picture was made more real than what it represents.
To conclude: Morris is tilting at windmills. Why? Because, at least to my mind, we have been moving steadily away from this basic set of assumptions in the last ten or twenty years. Rather than worry about abstract systems, language, and signifiers, we have become much more concerned with things, practices, and experiences themselves. Ian Hacking’s wonderful book Representing and Intervening is an excellent early example. Hacking was a philosopher of science, but I think there is a really important trend that generalized to historians as well. We have begun to write more about material culture and laboratory practices than language or hegemonic political structures. Similarly, philosophers of science these days rarely still try to solve logical puzzles about black and white ravens. Rather, they tend to look in detail at what scientists actually do. What I see, then, is a fundamental move from large-scale epistemic questions to more local ones (which often have a more ontological flavor). I don’t mean this in a shallow way, because of course Carnap himself was very concerned with ontology. In a sense I do mean it literally though, in a very flat-footed way. Let me pose it as a question: Have we indeed become more interested in the world rather than abstract structures representing it? If so, Morris may have a point, but he’s at least ten years too late for it to make much of a difference.
