The thrust of Liberal Naturalism, as it has been presented thus far, has been that phenomenal consciousness can only be satisfactorily accounted for by introducing some new, fundamental, nonphysical aspects to our theories of nature. In the process of constructing such new fundamental theories, however, we may find some surprising or counterintuitive results. In particular, we might end up at a panexperientialist theory. Panexperientialism is the view that experience outruns cognition-- that experience does not belong only to cognitive systems like human brains. (Panexperientialism is not to be confused with panpsychism; the latter holds that minds are more ubiquitous in nature than we might think, while the former only holds that some form of subjective experience is more ubiquitous than we think. The 'ubiquitous' subjective experience posited by panexperientialism need not be nestled in thinking, cognizing minds.) Given that infusing our theories of nature with new, fundamental aspects to account for consciousness in the human case may very well lead us to similarly view phenomenal consciousness itself as a rather ubiquitous fundamental aspect of nature, we need to verify that panexperientialism is even a viable option to begin with. Is it a coherent view? Does it withstand criticism? In this chapter, Rosenberg investigates possible critiques against panexperientialism and ultimately finds it to be a coherent theoretical possibility. Before he begins this discussion, he introduces a new term: "qualitative field." A qualitative field is a bounded collective of experienced phenomenal qualities. E.g., my subjective experience is a qualitative field, as distinct from the qualitative field that constitutes your subjective experience. So this is essentially another term for subjective experience or phenomenal consciousness, although with a more explicit acknowledgement of the boundedness of experience (as discussed in chapter 4). Using this term will also make some discussions flow more naturally; for instance, we can ask "what is it that differentiates two qualitative fields?" instead of saying "what is it that differentiates the bounded p-consciousness of one system from that of another?" Is There Evidence for Panexperientialism? The first set of critiques against panexperientialism that Rosenberg addresses is 1) there is no evidence for p-consciousness outside of cognitive contexts, and 2) panexperientialism is incoherent because it supposes experiences without appropriate (cognitive) experiencers. Objection 1 is easily dispatched by noting that every theory of p-consciousness, save perhaps solipsism, goes beyond the direct evidence. Our only immediate evidence that subjective experience exists is in our own, first person cases. As a result, what we consider to constitute evidence for p-consciousness in cases other than our own depends heavily on theory. For instance, the basis for objection 2 is founded in the theoretical (not empirical) notion that only cognitive systems can support p-consciousness. Accordingly, if we can find a simpler, better motivated, and more coherent theory that implies that certain non-cognitive or otherwise surprising systems support experience, we should accept these consequences as discoveries about nature. Doing such is in accordance with the necessarily deep link between the nature of our theories of consciousness and what we count as evidence for p-consciousness in cases other than our own. Is Panexperientialism Coherent? Next, Rosenberg addresses the concern that the notion of experience existing outside of a cognitive context is incoherent. Can it really be that experience could exist without, in some sense, being essentially mental or cognitive? Rosenberg offers four observations aimed to dilute our confidence that all kinds of experience must be fundamentally mental. 1. Our concepts of experience are highly open ended. Because of the fundamentally bounded nature of p-consciousness, we cannot peer directly into other qualitative fields. We must, for instance, allow that a manta ray sensing electromagnetic radiation might have corresponding experiences that are completely foreign to us, just as a blind man must allow that those who can see have experiences that are completely foreign to him. Accordingly, we must allow that simpler and simpler beings might have more and more primitive forms of experience. As we move down this scale, there is no clear boundary at which point we are definitively forced to stop attributing experience. Our confidence may grow weaker for systems that increasingly do not resemble us, but weak confidence for a proposition does not amount to clear logical grounds for rejecting it outright. 2. It could be that there are certain types of experiences that are completely alien to what is presented in human experience. We do not have to suppose that every experiential event that occurs in a panexperientialist world is some diminished analogue of human experiences, like 'a little pain' or 'a little speck of blue.' Rather, the kinds of experiences that might obtain in non-cognitive systems could be so foreign from human experience as to be inconceivable for a human. The strange nature of such qualitative fields need not be limited to just their qualitative character; they could very likely be non-mental, and particularly non-intentional, in a way that would be essentially impossible for a human to fully empathize with. 3. The best way to conceive of a non-mental qualitative field is by way of analogy. Rather than try to directly conceive of what such qualitative fields Y belonging to systems X would be like, we can draw out their abstract resemblance to our own, human experience using the analogy, 'Y is to system X as experience is to the human mind.' In so doing, we directly acknowledge that we cannot fully understand such qualitative fields, since we cannot experience them firsthand and directly feel their qualitative character. However, we also maintain that we understand them in an abstract way, as standing in the same sort of relation to the systems to which they belong as our own subjective experiences stand in relation to our own brains/minds. 4. The best name for non-cognitive qualitative fields is protoconsciousness. This term is suggestive of the assertion that they are not being experienced by a cognitive mind, but instead are rather primitive states of 'pure experience.' Rosenberg concludes that observations 1-3 suggest that resistance to panexperientialism arises from cognitive rigidity rather than conceptual incoherence. It seems our concepts of experience can tolerate a panexperientialist expansion with falling into incoherence. Resistance to panexperientialism, then, seems to be grounded in biases formed from our own fixed positions as cognitive experiencers. We are so used to associating subjective experience with minds that it may make us uneasy to attribute experience to non-mental systems, but this unease is not grounded in purely logical concerns. Rosenberg also cautions that we need not assume that every conceptual object, such as a rock or thermostat, is a subject of experience on a panexperientialist view. Depending on the details of the theory, large scale systems that support bounded, experiencing individuals might be rare. It could be that a panexperientialist view only attributes qualitative fields to large scale systems such as the ones we would otherwise expect-- humans, some animals, some computers, etc.-- and only adds unexpected, non-cognitive subjects of experience at the smallest scales of nature. (Although such protoconscious phenomena would not have thinking minds, they would still count as subjects of experience insofar as they would be bounded qualitative fields.) Protoconsciousness and Rrepresentationalism Because protoconscious properties can exist outside of cognitive contexts according to panexperientialism, they might not have certain features that we normally associate with consciousness, particularly representational or intentional features. But on some views, representational features are essential to phenomenal properties, and thus phenomenal properties without representational content cannot exist. Call any variety of such a view representationalism. If representationalism is right, it follows that protoconscious properties cannot be phenomenal properties. Therefore, any panexperientialist view must answer to representationalism if it is to get off the ground. Rosenberg has three possible responses to the representationalist challenge against panexperientialism, each of which he regards as adequate. 1. Protoconscious properties do, in fact, have intrinsically representational features, but these are just not used unless in they are placed in the proper cognitive context. In this case, the representational features of protoconsciousness would be like the charge of an ion that is located in an environment where its charge cannot interact with other charges. On this view, there is no challenge to representaionalism, so it need not be argued against. 2. Representationalism is problematic for the human case, and thus we should not hold it to be a serious threat against panexperientialism. Rosenberg's main ammunition againt representationalism here comes from case studies of synesthesia. Synesthesia is a condition in which a person experiences various kinds of cross-modal perception; for instance, hearing a sound might invoke a visual sensation of color, or smelling an odor might invoke a tactile sensation of shape. Importantly, synesthetes can press their cross-modal perceptions into representational services. For instance, one synesthete named Carol, who experiences the visual color orange upon feeling pain, reports a story where she used her synesthetic perception of orange to diagnose nerve damage of one of her teeth at the dentist. Carol's experience suggests that the same representational content can be expressed with varying phenomenal contents, undermining the representationalist claim that representational contents determine phenomenal contents. Another synesthete, DS, experiences visual shapes upon hearing sounds, and conceives of the shapes as fundamentally constituting part of what it is to hear. DS's testimony seems to undermine a further representationalist claim that phenomenal contents determine representational contents. It is not clear that the shape aspect of DS's heard shapes represent anything about the sounds he hears, nor that they represent in a way consistent with the manner in which an analogous, 'normally perceived' visual shape represents a physical object. 3. Even if representationalism is true for human consciousness, there is no basis for generalization to non-cognitive cases. Generalizing from the human case to make conclusions about all possible types of phenomenal properties is problematic, because the human case makes for a highly biased sample set. Because we can only access first person information about consciousness in our own cases, and because our own cases are unfailingly instances of cognitive minds, we may be led to believe that all types of phenomenal properties must be representational-- but this conclusion suffers from the same methodological problems as concluding that (eg) all bears are white, upon making extensive observations of bears in the North Pole.