Is the United States Phenomenally Conscious? Reply to Kammerer
Department of Philosophy
University of California at Riverside
Riverside, CA 92521-0201
April 28, 2016
Is the United States Phenomenally Conscious? Reply to Kammerer
In my article “If Materialism Is True, the United States Is Probably Conscious” (Schwitzgebel 2015), I argue that the United States, considered as a concrete entity with people as some or all of its parts, meets all plausible materialistic criteria for phenomenal consciousness, such as self-monitoring, complex internal information processing, and sophisticated goal-directed behavior. François Kammerer (2015) offers a thoughtful critique, defending materialism against this unwanted conclusion by means of a sophisticated “anti-nesting principle”.
Anti-nesting principles are principles according to which conscious beings cannot have conscious subparts. Consciousness, such principles say, can only arise in one organizational level at a time. If an entity is conscious, then none of its subparts are conscious, nor is there any consciousness in any larger system to which that entity belongs. Conscious beings cannot in this sense “nest”. Restricted anti-nesting principles allow that conscious beings might sometimes nest, while specifying a set of restricted conditions under which they cannot nest. If an appropriate anti-nesting principle is correct, then on the assumption that individual citizens of the United States are conscious, the United States as a whole could not be conscious.
Putnam (1967) and Tononi (2012) have both prominently advocated unrestricted anti-nesting principles – which I critique in Schwitzgebel (2015) – but overall the literature on this topic is sparse. Kammerer offers the following restricted Sophisticated Anti-Nesting Principle:
Given a whole W that instantiates the functional property P, such that W’s instantiation of P is normally sufficient for W to instantiate the conscious mental state S, W does not instantiate S if W has at least one subpart that plays a role in its functional organization which fulfills at the same time the two following conditions:
(A) The performing of this role by the subpart requires (given the nature of this functional role and our theory of consciousness) that this subpart has conscious mental states (beliefs, emotions, hopes, experiences, desires, etc.) that represent W (what it is, what it does, what it should do). That is to say, this subpart has a functional property Q, Q being a sufficient condition for the subpart having the conscious mental state R (where R is a mental state representing W).
(B) If such a functional role (i.e., a functional role of such a kind that it requires that the subpart performing it has conscious mental states representing W) was not performed by at least one of the subparts of W, W would no longer have the property P (or any other functional property sufficient for the having of S). In other words: if no subpart of W had R, then W would no longer have S (p. 1051).
Short, hopefully-not-too-oversimplified version: If the reason a larger entity acts as if it’s conscious is that it contains smaller entities within it who have conscious representations of that larger entity, then that larger entity is not in fact conscious.
One possibly attractive feature of Kammerer’s view is this: It permits entities to contain conscious subparts in some intuitively plausible cases, such as my “Antarean antheads” – mobile ant hives that exhibit highly sophisticated behavior at a group level but insect-like intelligence at the level of individual ants; and also in Ned Block’s (1978/2007) case in which a traveler inhales a tiny conscious organism whose behavior is incorporated into the normal functioning of her brain.
Another possibly attractive feature of Kammerer’s view is this: There’s something intuitively plausible about denying consciousness to an entity whose behavior depends excessively, in the wrong way, on a conscious entity inside who is directing it – a car-and-driver system, for example. (Presumably the driver is conscious but the car-and-driver system is not also conscious at a higher level.) Both Dretske’s and Chalmers’s objections to U.S. consciousness, as I articulated those objections in my original essay, are compatible with versions of this thought. Kammerer’s version has the virtue of avoiding the concerns I raise about the Dretske and Chalmers objections while retaining some of their spirit.
One possibly unattractive feature of Kammerer’s view is that it gives a major role to the conscious representation of complex group entities, which one might resist if one either rejects complex cognitive phenomenology (see Bayne and Montague 2011) or regards conscious representations as functionally irrelevant to important cognitive tasks (Moore 2016).
I have three main concerns about Kammerer’s Sophisticated Anti-Nesting Principle.
First: Kammerer’s principle is compatible, as he acknowledges, with the existence of group-level phenomenal consciousness, including possibly in some actual or close-to-actual conditions.
This isn’t so much a concern about the principle itself as it is a concern about Kammerer’s deployment of that principle as an objection to my argument. Although the title and official thesis of my essay is “If Materialism Is True, the United States is Probably Conscious”, the main point that I am really after is that if materialism is true, some actual (or easily arranged) groups of people are probably literally phenomenally conscious. I choose the United States as my test case not because of any particular interest in the U.S. in particular, but rather because it seems the most plausible case to make. If Kammerer would allow – as his argument seems to – that the United States is not a good case but there are other actually existing conscious entities that contain people as subparts, then Kammerer and I can join in agreement on my intended main idea, disagreeing only about a detail. (I doubt that Kammerer would in fact allow that there are actual or nearby cases of group consciousness, but nothing he says seems to preclude this possibility.)
Kammerer’s principle might even allow the actual group-level phenomenal consciousness of “the United States” in my intended usage of the term. Whether Kammerer’s principle would allow this appears to depend on some tricky issues about representation, reference, and the ontology of groups. If people don’t have conscious mental states that represent the entity that I am referring to with the term “the United States”, then Kammerer’s anti-nesting conditions will not be satisfied and the consciousness of this entity will not be excluded. To settle this point requires asking: Do people have conscious mental states that represent “the United States” in the relevant sense of the term?
In the original paper, I defined “the United States” as “a spatially distributed, concrete entity with people as some or all of its parts” (Schwitzgebel 2015, p. 1698 and 1708). If Kammerer’s new anti-nesting principle is to apply, then people must have representations of that entity. But it’s not clear that they do. When people use the term “the United States” they might refer to the United States government, or they might refer to the nation, or they might refer to the land mass plus possibly the people on it, or maybe something else. My best guess is that the most helpful way to conceive of “the United States” for purposes of the group consciousness question is as a vague-boundaried entity that includes all the citizens and residents of the United States, plus their artifacts (especially their computers, and maybe also other household goods and even roads), and maybe the air between people, but not including the terrain or unharvested crops. I’m not at all committed to that particular way of conceptualizing the United States, but I think it’s one reasonable way to divide internal cognition (such as messages traded via computer) from outside resource extraction (such as farming and mining). The question is then whether people succeed in consciously representing that thing, the thing I am claiming is probably conscious if materialism is true, rather than some other ontologically related but different thing, when they talk and think about “the United States”. It’s not clear. Maybe if it’s a good enough natural or artificial kind, it will be a “reference magnet” for people’s representations (though since reference magnetism is often treated as competitive it would have to be a better magnet than the various competitors). That seems a rather risky ontological-cum-referential proposition. And if people do not consciously represent “the United States” in the relevant manner, then Kammerer’s anti-nesting conditions are not met.
Second: Kammerer’s principle appears to make the consciousness of the group depend in an unintuitive way on the motivations of its members. Now it’s my view that all theories in this area will have some sharply unintuitive consequences (see also Schwitzgebel 2014), so this isn’t a decisive objection. But if part of the attraction of Kammerer’s principle is that it appears to accord with intuition, then any unintuitive consequences should be clearly noted and weighed up on the negative side of the ledger before deciding whether to accept the principle.
According to Kammerer’s Condition A the subpart (the person) must not only consciously represent the whole, but that representation must also be required for the subpart to do its bit. The intuitive point here is that if a bunch of conscious beings are already doing what’s necessary to create group-level consciousness and then they discover that they are doing so, and thus come to represent the whole at the group level, but nonetheless still continue doing what they’ve always done, this should not cause consciousness at the group level to suddenly cease. They represent the whole, now, but that representation is not required for them to continue doing what they’ve always done to bring about that group-level conscious experience.
“Required” is a tricky word. It’s hard to know exactly how to read it, or how far afield to go in thinking about counterfactual cases, but on one possible flat-footed reading, proper motivation can sometimes be “required” for a person to do something. So let’s suppose that Leo is a member of a conscious group and he is responsible for one important task – a task, perhaps, essential to the group entity’s “seeing” something in some region of its environment. Maybe he manages the video feeds from Sector 27A of the Tonga Trench. On Monday, he did not know about or represent the conscious group to which he belongs; he just did his business, making his contribution. On Tuesday, he learns about the existence of this conscious group, and now he does represent it, but he keeps on doing his thing. So far it seems plausible to say that his representation of the group is not “required” for his contribution to the group-level processes in virtue of which the group is conscious. So on Tuesday Condition A continues not to be met and thus the group remains visually conscious of Section 27A of the Tonga Trench. On Wednesday, however, Leo becomes depressed. He momentarily decides that he’s going to stop processing the video feeds. But then he thinks to himself that stopping would cause the group to lose visual consciousness of that sector of the Tonga Trench. On Monday such considerations wouldn’t have occurred to him, but now that he represents the larger group as a group he feels obligated to continue processing the feeds. No one else in the group is aware of Leo’s hesitation, and it has no impact on how he processes the feeds. Condition A is arguably met: Leo’s representation of the group is (motivationally, emotionally) required for him to contribute his part to the group’s cognition. Thus, applying Condition A, now the group is not visually conscious of Section 27A of the Tonga Trench. This, despite presumably no difference in group-level self-report or behavior with regard to the Tonga Trench. Now suppose on Thursday Leo feels better, and the group level representation is no longer required for him to feel motivated to do his task. Consciousness would seem to be restored, by Kammerer’s criteria as I am currently interpreting them. Again, there need be no change in the behavior or processing of the system as a whole. This seems unintuitive.
I have run the case in terms of emotional requirement, but I don’t think the worry depends on casting it in exactly that way. The general form of my worry is this: If a person who is a subpart of a larger conscious organism represents the conscious whole, then depending on how one interprets the idea of “requirement”, there will likely be possible cases where that representation is at first not required for the person to participate and then something in the background conditions changes so that the representation later becomes required – with no difference to the processing of information by the whole. In such cases, by flipping back and forth across whatever the dividing line is between “required” and “not required”, it should be possible to construct bizarre flickering-qualia cases in which conscious experience at the group level flips between present and absent despite a lack of any group-level functional change (cf. Chalmers 1996).
Kammerer could resist the Leo case by raising the bar for “requirement”. Since Leo was doing his job on Monday, clearly there is some interpretation of “required” on which even on Wednesday his conscious representation is not required for him or someone like him to play the relevant functional role. Taken to an extreme, raising the bar for “requirement” in this way threatens to render the Condition A nearly impossible to meet. For example, applying Kammerer’s own criteria, we might imagine the population of the United States being replaced by non-conscious “people” who behave functionally identically to the current population but who fail to consciously represent the U.S. as a whole because they in turn are constituted by lower-level people. Kammerer allows that complex entities can implement complex functional roles, roles normally sufficient to instantiate consciousness, without themselves instantiating consciousness because some different but related role is being played by a conscious entity at a lower level. It seems to follow that consciousness cannot be “required” of any particular functional state for a person to do her part in the functioning of the U.S., if requirement is interpreted with a very high bar.
If it’s too easy for a conscious representation to be “required”, Kammerer runs into the Leo case; if it’s too hard, his anti-nesting conditions risk being impossible to meet because his own theory allows complex functional roles to be filled nonconsciously. It’s not clear where between these two extremes Kammerer would want to drive the stake. Regardless of where the stake is driven, any theory that dissociates higher-level functional organization from the presence or absence of phenomenal consciousness would appear to invite some variant of flickering or fading qualia cases, if the purported conditions for consciousness can be manipulated without changing higher-level functional organization.
Third: Kammerer suggests that his anti-nesting principle is well motivated and non-ad-hoc, but I’m not sure that this is so. Crucial to Kammerer’s non-ad-hocness claim is the idea that
In particular, one should only ascribe consciousness to an entity when one cannot explain the behavior and the organization that seems to justify this ascription as the consequence of mental states of other, distinct subjects – notably mental states of other subjects which bear on the very behavior and organization of the entity…. Such a principle can itself be justified (even though a satisfying justification would require further reasoning) by appealing to the idea that the ascription of consciousness is, amongst other things, supposed to play a role in the explanation and justification of the behavior of the entities to which it is ascribed. But, as for all explanations and justifications, one should always choose the simpler explanation/justification when faced with many explanations/justifications (p. 1055).
One problem with this general type of reasoning is that it risks driving us headlong into the overdetermination or causal exclusion problem, which has been a substantial source of trouble both in philosophy of mind (esp. Kim 1993) and in the ontology of objects (esp. Merricks 2001; Thomasson 2007). If Kammerer’s reasoning here is intended to apply generally, then it raises questions about whether even normal human consciousness exists (since arguably human behavior could in principle be explained at the cellular or molecular level) or whether baseballs break windows (since arguably the breakage is explainable at the lower level by appealing to the behavior of many individual particles arranged baseballwise). One common answer to such concerns is to allow that higher level states, events, or entities nonetheless exist or have causal/explanatory power – maybe because of “emergent properties” or maybe without any type of special emergence. But if this kind of move is available in general for defenders of baseballs and human consciousness, then it’s not clear why a parallel move wouldn’t be available to the defender of group consciousness.
Alternatively, Kammerer might intend his claim not to depend on a general simplicity principle that applies also to human mental state explanation and baseball causation. But now the principle seems unmotivated. Why commit to a specific simplicity principle that applies only to the relationship between consciousness and other mental states if one doesn’t wish to commit to similar principles of a more general form? Without a clearer explanation of the appeal of such a specific version of simplicity principle, I worry that Kammerer’s principle is ad hoc or question-begging after all – appealing to us only because it seems to rule out U.S. consciousness.
A second concern about Kammerer’s application of a simplicity principle here is that group-level explanation is often arguably simpler than individual level explanation. Why did the army disembark here rather than there? Because that gave them the clearest path to the capital. An explanation in terms of individual generals’, sergeants’, and privates’ motives would be vastly more complicated. So if “one should always choose the simpler explanation/justification when faced with many explanations/justifications”, then there might well be reason to prefer group-level psychological explanations for group-level behavior.
I have not encountered a better defense of an anti-nesting principle than Kammerer’s defense here. Anti-nesting principles ought to have considerable appeal as a means of avoiding unintuitive consequences regarding both group consciousness and consciousness at organizational levels lower than the individual organism. The difficulties that arise reveal how far we still have to go in thinking through this relatively neglected but fundamentally important issue for theories of consciousness.
Bayne, Tim, and Michelle Montague, eds. (2011). Cognitive phenomenology. Oxford: Oxford.
Block, Ned (1978/2007). Troubles with functionalism. In N. Block, Consciousness, function, and representation. Cambridge, MA: MIT.
Chalmers, David J. (1996). The conscious mind. Oxford: Oxford.
Kammerer, François (2015). How a materialist can deny that the United States is probably conscious – response to Schwitzgebel. Philosophia 43: 1047-1057.
Kim, Jaegwon (1993). Supervenience and mind. Cambridge: Cambridge.
Merricks, Trenton (2001). Objects and persons. Oxford: Oxford.
Moore, Alan T. (2016). The experience of reading. PhD dissertation, Department of Philosophy, University of California at Riverside.
Thomasson, Amie L. (2007). Ordinary objects. Oxford: Oxford.
Schwitzgebel, Eric (2014). The crazyist metaphysics of mind. Australasian Journal of Philosophy 92: 665-682.
Schwitzgebel, Eric (2015). If materialism is true, the United States is probably conscious. Philosophical Studies 172: 1697-1721.
 For helpful discussion, thanks to David Holiday (esp. regarding the car and driver system), François Kammerer, Jeremy Pober, and commenters on The Splintered Mind blog.