Simple vs Sophisticated Empiricism
I here refer to "Hume Shifts the Burden of Proof", by Kelley L. Ross.
In this essay, Ross argues persuasively against Hume's version of empiricism. I do not want to use a word as harsh as "crude" to describe the empiricism of Hume, although it is certainly very crude compared to the various versions of empiricism available today. However, Hume existed before the days of knowledge of proteins, vitamins, carbohydrates, minerals, electromagnetism and atomic, nuclear, and particle physics. It therefore seems charitable to call his empiricism "simple" rather than crude, if that is not too condescending.
The basic problem with Hume's version of empiricism is that it is far too restrictive on the empirical validity of concepts. In order to be empirically valid, that concept must correspond to an atecedent sense impression. In the case of microbes, which one cannot see by unaided observation, but can see through the aid of a microscope, one might still claim that microbes are thus reducible to sense impressions. But then what are we to make of quantum mechanics, which posits entities that are unobservable in principle? It seems strange to deny that their corresponding concepts thereby lack empirical validity. I would argue that this is because such concepts constitute the best explanation for the sense impressions that we do possess. After all, it would be arbitrary to suppose that the best or simplest possible explanation for our sense impressions had itself to reduce to a sense impression.
In this way, we can see that a more sophisticated version of empiricism is possible. The basic project of the reduction of concepts to sense impressions is still retained, but it is no longer necessary for every single concept to be reducible in such a way. If some concepts are deemed optimal for the explanation of other concepts that do reduce to sense impressions, then this is good enough.
Certainly in modern day physics, we have many available examples of concepts that are irreducible to sense impressions, such as quarks. However, it is important to note that even in Hume's day, some irreducible concepts still existed. Neither power, force, energy nor necessary connection are reducible to sense impressions, as he realised, and they were clearly indispensible to the explanations of physics, then as now. Yet he thought that he was preserving these concepts for science by simply claiming that we believed in them out of "custom", or "habit". Yet this has the obvious difficulty that they cease to tell us anything about the world, only about our own minds, as Ross notes in the essay.
One must, therefore, wonder how Hume could have been content with such a skeptical conclusion. Ross argues that the reason is simply that, as the title of the essay states, he shifts the burden of proof. That is, he starts out by asserting that the only meaningful concepts are those that reduce to sense impressions. His argument for this is on the grounds that if anyone can produce for him an example of a meaningful concept, then it becomes his responsibility to demonstrate the sense impression to which it corresponds. Unfortunately, he does not adhere to this principle consistently. In fact, almost immediately, he takes it back and shifts the burden of proof, arguing that in order for someone to argue that a concept is meaningful, she must supply a sense impression to which it corresponds.
This sleight-of-hand enables him to argue for his simple empiricism in a completely circular manner. He has not provided any evidence whatsoever that in order for a concept to be meaningful, it must be shown to be reducible to a sense impression. All he has done is offer to produce such a sense impression for concepts offered as counterexamples, and then later on simply to repudiate such concepts as meaningless. Therefore, a large basis for much of the strong skepticism that appears later in his work has already been undermined.
In this essay, Ross argues persuasively against Hume's version of empiricism. I do not want to use a word as harsh as "crude" to describe the empiricism of Hume, although it is certainly very crude compared to the various versions of empiricism available today. However, Hume existed before the days of knowledge of proteins, vitamins, carbohydrates, minerals, electromagnetism and atomic, nuclear, and particle physics. It therefore seems charitable to call his empiricism "simple" rather than crude, if that is not too condescending.
The basic problem with Hume's version of empiricism is that it is far too restrictive on the empirical validity of concepts. In order to be empirically valid, that concept must correspond to an atecedent sense impression. In the case of microbes, which one cannot see by unaided observation, but can see through the aid of a microscope, one might still claim that microbes are thus reducible to sense impressions. But then what are we to make of quantum mechanics, which posits entities that are unobservable in principle? It seems strange to deny that their corresponding concepts thereby lack empirical validity. I would argue that this is because such concepts constitute the best explanation for the sense impressions that we do possess. After all, it would be arbitrary to suppose that the best or simplest possible explanation for our sense impressions had itself to reduce to a sense impression.
In this way, we can see that a more sophisticated version of empiricism is possible. The basic project of the reduction of concepts to sense impressions is still retained, but it is no longer necessary for every single concept to be reducible in such a way. If some concepts are deemed optimal for the explanation of other concepts that do reduce to sense impressions, then this is good enough.
Certainly in modern day physics, we have many available examples of concepts that are irreducible to sense impressions, such as quarks. However, it is important to note that even in Hume's day, some irreducible concepts still existed. Neither power, force, energy nor necessary connection are reducible to sense impressions, as he realised, and they were clearly indispensible to the explanations of physics, then as now. Yet he thought that he was preserving these concepts for science by simply claiming that we believed in them out of "custom", or "habit". Yet this has the obvious difficulty that they cease to tell us anything about the world, only about our own minds, as Ross notes in the essay.
One must, therefore, wonder how Hume could have been content with such a skeptical conclusion. Ross argues that the reason is simply that, as the title of the essay states, he shifts the burden of proof. That is, he starts out by asserting that the only meaningful concepts are those that reduce to sense impressions. His argument for this is on the grounds that if anyone can produce for him an example of a meaningful concept, then it becomes his responsibility to demonstrate the sense impression to which it corresponds. Unfortunately, he does not adhere to this principle consistently. In fact, almost immediately, he takes it back and shifts the burden of proof, arguing that in order for someone to argue that a concept is meaningful, she must supply a sense impression to which it corresponds.
This sleight-of-hand enables him to argue for his simple empiricism in a completely circular manner. He has not provided any evidence whatsoever that in order for a concept to be meaningful, it must be shown to be reducible to a sense impression. All he has done is offer to produce such a sense impression for concepts offered as counterexamples, and then later on simply to repudiate such concepts as meaningless. Therefore, a large basis for much of the strong skepticism that appears later in his work has already been undermined.
Comments