Searle defends the view that there is no necessary connection between consciousness and behavior. He also argues that consciousness is essentially a first-person phenomenon, and criticizes the "problem of other minds." Finally, he distinguishes between three kinds of intentionality. Indented paragraphs are my comments.
1. Consciousness and behavior
Three versions of a thought-experiment in which the brain is slowly replaced by silicon chips. (a) conscious experience remains the same and behavior remains the same. (b) conscious experience gradually disappears, although behavior remains the same. [note that this might be regarded as a version of the "zombies" or "absent qualia" objection to functionalism.] (c) consciousness remains the same, but behavior gradually becomes impossible.
That (b) and (c) are conceptual possibilities is designed to show that there is no conceptual link between behavior and conscious states.
Do (b) and (c) actually show anything this strong? If they are possible, then behaviorism can't be true, that is, conscious states can't just be identified with behaviors. But for a functionalist the link, although necessary, is less direct. Compare: the heart's function is to pump the blood, but if it is removed and placed in a jar it won't pump blood any more. Maybe case (c) is rather like that.
Note that the argument has to do pretty exclusively with conscious states. Searle argues that this is not much of a limitation, since in chapter 7 he will defend the view that unconscious states must be potentially conscious. But if we don't buy this, we may think the arguments don't show much about intentional states such as beliefs or desires. And don't belief and desire seem more closely linked to behavior than consciousness? One way to put it: what are beliefs and desires for if not to guide action?
2. First-person character of consciousness
If (b) and (c) are possible, then it's possible for there to be conscious states that there is absolutely no way to learn of from a third-person perspective (case c), and also it's possible for it to appear, for all one can tell from a third-person point of view, that a subject is conscious even though in fact the subject is not conscious (case b).
3. Other minds
Searle makes two points: (1) our evidence for consciousness in others is not their behavior alone, but rather a combination of facts about their behavior with facts about their neurophysiology. (It's the biology that's essential; the relevance of behavior is due only to the fact that it's a clue to how it was caused.) (2) In a sense there isn't a problem of other minds at all, because for most of us, most of the time, there is no such explicit hypothesis. I just act in certain ways toward other people (part of the Background).
Does the fact that an issue arises only in philosophy mean it's not a real issue? No doubt Searle is right that "the question doesn't arise except in philosophical debate." So what? If our Background practices only make sense on the assumption that other people are conscious, then it seems meaningful to raise the issue of how we can know this, even if it's an issue that normally "doesn't arise."
Intentionality can be intrinsic (I'm thirsty, I believe it's 78 degrees, etc.), "as-if" (the lawn is thirsty, the thermostat believes it's 78 degrees), or derived (The sentence "I'm thirsty" means that I am thirsty, the "78" mark on the thermostat represents a temperature of 78 degrees).
By 'intrinsic' Searle means "the real thing" as opposed to make-believe or derived. (In this sense 'intrinsic' is not opposed to 'relational'.)