Philonous now resumes the task of demonstrating that we do not mediately perceive mind-independent material objects — that is, that we have no reason to infer their existence from our immediate experience of the world. He has already demonstrated, he believes, that we do not infer matter as a substratum, or support for observable qualities, and now he will try to show that we cannot infer material objects as the archetypes for our ideas. In the second dialogue he will consider the third possibility for an inference to mind- independent material objects: material objects as the cause for our ideas.
Hylas, of course, is the one who introduces the idea of material objects as the archetypes for our ideas. His hypothesis is familiar enough: our ideas are copies of things out in the world; they resemble objects out in the world and so when we perceive our ideas we also gain access to the material objects which they resemble. Ideas are related to material objects, in other words, like a photograph is related to the person photographed. Our immediate access is to the photograph, but through it we gain access to the person herself. In the case of ideas, it would go like this: there is a tree out in the world, and then there is an idea of a tree, which is just a copy of that tree out in the world (like a snapshot). What we perceive is the copy, but through that copy we come to know about the original tree as well.
In order to debunk this notion of material objects as the archetypes for our ideas, Philonous presses of the notion of representation. He says that it is crazy to think that an idea can represent a mind-independent material object. Only an idea, he tells Hylas, can resemble an idea. Take a materially existing tree and the idea of a tree. How can these two things resemble one another? The idea of the tree is green and brown. But the material tree cannot be green and brown because colors, he has already shown, cannot exist outside of the mind. The idea is large and tree-shaped. But the material tree cannot have either of these properties either, because they, too, are mind-dependent. In short, our ideas of things (trees, flowers, chairs, etc.) are completely comprised of sensible qualities and no sensible qualities can exist outside of the mind. So whatever exists outside of the mind, cannot in any way resemble our ideas. Ideas are sensible, and real things are not. To say they resemble one another would be like saying that something invisible resembles something colored. In addition, our ideas are fleeting and variable, while material objects are supposed to be stable and constant. In this way, too, they cannot resemble one another.
Hylas is now finally reduced to skepticism. He admits that no sensible things exist outside of the mind, and concludes from there that no sensible things have any real existence. Denial of the real existence of sensible things, remember, was precisely our definition of skepticism. So Hylas' materialism really has led him straight into skepticism. In the next dialogue, Philonous will demonstrate how idealism can pull him out of this skepticism.
Since all of us are probably just as reluctant as Hylas to accept Berkeley's idealist picture, it is worthwhile to analyze how we could be materialists and still avoid skepticism. Is Berkeley really right in claiming that materialism leads us straight into those pervasive doubts? Other philosophers have effectively avoided this snare. In particular, we can look at how Rene Descartes and John Locke avoided skepticism. Both of them had to deal with these worries because both of them ascribed to a mediated theory of perception (i.e. a theory of perception on which our immediate access is to ideas, not objects).
Both Descartes and Locke agree that we must infer the existence of mind- independent material objects; proof of their existence is not given to us immediately in experience. Descartes infers the existence of mind-independent material objects from his innate ideas. In particular, he uses his innate idea of God as a perfect being to establish that God could not be a deceiver, and, from there, reasons that if our sensations were caused by anything other than mind-independent material objects God would be deceiving us. We will see later that Philonous actually considers a similar line of reasoning, and then rejects it.
Locke, as an empiricist, cannot infer the existence of mind-independent material objects from innate concepts; he must infer them from his sensory experience. In fact, though, Locke has three strategies for dealing with this concern, and he employs all of them in chapter xi of the Essay Concerning Human Understanding. Locke's first strategy, and the one he seems most viscerally drawn to, is to simply refuse to take the skeptic seriously. Can anyone really doubt, he asks, that there is an external world out there? Another strategy he makes use of is to give a pragmatic response. If you want to doubt that there is an external world, he says, that is just fine. All that matters is that we know enough to enable us to get around in the world.
Throughout the chapter, Locke formulates a long and detailed argument based on inference to the best explanation. He presents a number of puzzling facts about our experience that can all best be explained by positing that there is an external world that is causing our ideas. Taken singly each one makes it a little more likely that there is an external world out there, but taken as a whole, Locke feels, they provide overwhelming evidence—so overwhelming that the inference is almost strong enough to be called knowledge. Locke brings up seven marks of our experience that can best be explained by positing an external world. The first is that there is a certain vivacity to perception that cannot be found, say, in memories or the products of the imagination. Berkeley too, as we will see, makes use of this mark of sensations. In chapter XI Locke offers six more empirical marks which distinguish this same set of ideas. He points our that we cannot obtain these ideas if we do not have the organ appropriate to them. No one born without the ability to hear, for example, can possibly have the idea of the sound of a french horn. Next Locke points out that we are able to receive ideas of this sort only in certain situations. Though the organs remain constant, the possibility of experiences change. It cannot, therefore, be the organs themselves that are responsible for producing these ideas. In section five Locke discusses the passive nature of these ideas; they are not voluntary, but involuntary, coming to us unbidden and unavoidably. We cannot just choose to have the experience of tasting watermelon at will, for example. Nor can we choose to avoid hearing the blaring siren of a car alarm at four o'clock in the morning.
The next empirical mark Locke brings forth involves pleasure and pain. Some ideas, Locke claims, cannot help but be followed by pleasure or pain. For example, when we have the sensation of seeing our flesh cut with a knife, this will almost certainly be accompanied by a sensation of searing pain (unless we are heavily medicated). When we call up the memory of these ideas, however, there is no experience of pain or pleasure accompanying them. In section seven Locke points out yet another empirical feature: a certain subset of our ideas fit into a coherent pattern, so that if we have one idea, we can, with great reliability, predict another one. The knife and pain example above can serve to illustrate this point as well. Another example of this mark of experience would be the fact that our sensation of seeing a hand let go of a book in midair is always followed by a sensation of seeing the book fall. Finally, not only is there a predictable correlation between the ideas of taste, vision, touch, sound etc. but there is also a correlation between the ideas belonging to different experiencing subjects (that is, between different people).
No single one of these marks proves conclusively that our experiences are caused by mind-independent material objects. However, as Locke points out, all of these marks, singly and as a group, can be coherently and compellingly explained by positing that our experiences are caused by mind-independent material objects. This makes that hypothesis overwhelmingly plausible — so much so, that it would be unreasonable for us to doubt it.
Berkeley never considers the possibility of proving the existence of mind- independent material objects by inference to the best explanation, but it is easy enough to guess what he would say about this line of reasoning. He would claim that his own idealist hypothesis explains all the evidence just as well as the materialist hypothesis. Each of those marks of experience, both singly and as a group, can be easily accounted for on his theory.
How could Locke then respond to Berkeley? He could counter that Berkeley's hypothesis does not explain the evidence just as well as the materialist hypothesis. For one thing, we tend to think that an explanation is better if it is simpler. But Berkeley's explanation is needlessly complex: where Locke just needs there to be objects out in the world, Berkeley needs there to be both God and the ideas he is causing us to have. In addition, where Locke just needs us to passively perceive the objects in order to have our experiences, Berkeley needs to tell a complicated story (as we will see) about how God shows us the ideas in his mind, and when He does this, and why. Furthermore, Berkeley does not even really succeed in accounting for all of the marks of experience. He never really explains, for instance, why our sensations always follow certain patterns, other than to assert that they follow these patterns because God shows them to us in these patterns. But why, we could press him, does God show us ideas in these patterns? Certainly, he is not bound by any physical necessity. Locke, the materialist, has a ready and satisfying explanation for why our sensations follow certain patterns: these are the patterns which the objects themselves are governed by, necessitated by physical law.
Locke, it seems, is on very good ground against Berkeley here. His materialist hypothesis explains these various marks of experience extremely well, whereas Berkeley's hypothesis explains the evidence to a lesser degree. Inference to the best explanation, then, favors materialism over idealism, which means, in short, that contrary to what Berkeley claims, we do have an excellent basis on which to infer the existence of mind-independent material objects. Materialism does not have to lead to skepticism, after all.