Hylas is still not ready to give up. He accepts that God is the ultimate cause of all our ideas, but wonders, could there not still be such a thing as matter? He believes that God may cause our ideas through the use of matter. This returns us to the discussion of whether we can infer the existence of mind-independent material objects. Hylas is now raising the final possibility: that we can infer matter as the cause of our ideas. He will raise this same possibility in several guises before he finally gives it up.
The idea that matter causes our ideas is familiar. Through the motion of particles, goes the usual theory, matter somehow stimulates us and gives rise to our ideas. Philonous, of course, does not put very much stock in this theory. Motion, he points out, is an idea, as we have already established. Ideas, though, are passive and inert and so cannot cause anything. The only thing in the world that is active is will, so will is the only causal agent. Only volition can act as cause. So matter cannot be the cause of our ideas.
But perhaps, Hylas presses, God uses matter as his instrument, in order to cause ideas in us. Philonous likes this idea even less than the previous one. First of all, he asks, what could these instruments be like, since we know that they cannot have any sensible qualities? More importantly, though, why on earth would God need an instrument? We only need an instrument for something if we cannot do it on our own, through a sheer act of the will, but God can do everything by a sheer act of the will. He's omnipotent. So God would never need an instrument.
Hylas is becoming increasingly desperate, and he suggests that matter is an "occasion", by which he means "an inactive, unthinking being, at the presence whereof God excites ideas in our minds". This would explain the order and regularity of our sensations — why we always have certain sensations followed by others, like the sensation of seeing fire followed by the sensation of feeling heat. Philonous points out that we already have sufficient explanation for this phenomenon in the fact of God's wisdom — this is just how He arranged things. And it is insulting to God to insinuate that he needs extra help from inert objects.
Hylas has one last possibility up his sleeve: matter is not a substratum, or an archetype, or a cause, or an instrument, or an occasion, but, rather, it is just something totally unknown, the abstract general idea of entity. In response to this suggestion, Philonous points out that this entity, whatever it is, cannot exist anywhere, because if it exists in space then it must exist in the mind since space or extension only exists in the mind. Then he asks whether Hylas has any positive notion of this entity at all, and Hylas admits he that he does not. Philonous concludes that this notion is, therefore, meaningless and empty, just as are all abstract general ideas.
Exhausted and defeated, Hylas admits that he has no way to answer any of Philonous' arguments. He stubbornly maintains, however, that he is unconvinced of idealism in his gut. Philonous counters that the only reason Hylas remains unconvinced, is that he is prejudiced against idealism. This does not indicate any weakness on the part of the arguments, but only the strength of Hylas' brainwashing.
When Locke makes his inference to materialism as the best explanation, the inference is one about causation. The materialist inference that he thinks the evidence warrants is this: the reason that sensory experience evinces the marks it does, it because sensory experience is caused by mind-independent material objects. This might help explain why Berkeley is so intent on attacking the possibility that we can infer that our ideas are caused by mind-independent material objects.
In the Principles, Berkeley devotes even more energy to deflating this type of inference, and putting his arguments there and his arguments here together we can come up with a detailed, step by step, argument against mind- independent material objects as the cause of our ideas. In both works, Berkeley imagines three causal scenarios for the production of our ideas. On the first of these, mind-independent objects out in the world cause our ideas. This, of course, is the basic materialist line. On the second scenario, our ideas are caused directly by God. This is his own view. Finally, the third possibility is that God causes our ideas through the medium of mind-independent material objects. This is the line that Hylas tries to push once he has given up the straight materialist line.
In all, Berkeley has five considerations that he thinks show that the second of these scenarios is the best. The first four of these considerations are intended to work against the first scenario, and the fifth is intended to discredit the last scenario. Against the first scenario Berkeley has this to say: First, even if our ideas were caused by mind-independent material objects there would be no way to verify whether this were true or not, because there is no way to get out of our own ideas and check. When we, for instance, see something with our eyes, we can check whether our perception was correct by touching the object, or tasting it, and so on. But we have no analogous way to check on all of our sense faculties; we have nothing else to go one, to check these as a whole. So we have no way to ever determine whether mind-independent material objects cause our ideas.
Second, this scenario is actually completely meaningless, because we cannot attach any meaning to our idea of mind-independent material objects. The only way to form a meaningful idea, according to Berkeley, is to conjure up a precise image. And the only way to conjure up a precise image, according to anyone, would involve ascribing sensible qualities. We cannot have an image of something that has no color, size, shape, and so on. But Berkeley thinks he has already shown that all sensible qualities are mind-dependent, and so cannot belong to mind-independent material objects. So there is no way to conjure up a meaningful idea of a mind-independent material object, which means that this scenario too is utterly meaningless.
Third, even allowing that the concept of mind-independent material objects is not meaningless, they still could never cause our ideas. According to the materialists themselves, mind-independent material objects are inert, and are moved around by some sort of force, like energy or God. Inert things, though, cannot be causal agents. How could something inert cause anything?
Finally, even allowing that mind-independent material objects could be causal agents, what is even more inconceivable is that material objects could causally interact with immaterial objects. The notion is completely incoherent.
Turning then to the last causal scenario, Berkeley simply maintains that he has a simpler account of what causes our sensory experiences. If God could cause our ideas directly, why would He go the intermediate route and cause them through inert immaterial objects that we have no access to anyway? There would be no point; from inside our experience, we cannot even tell the difference — it would look the same to us no matter which of these causal scenarios was the true one.