The Nesting Problem for Theories of Consciousness

In 2016, Tomer Fekete, Cees Van Leeuwen, and Shimon Edelman articulated a general problem for computational theories of consciousness, which they called the Boundary Problem. The problem extends to most mainstream functional or biological theories of consciousness, and I will call it the Nesting Problem.

Consider your favorite functional, biological, informational, or computational criterion of consciousness, criterion C. When a system has C, that system is, according to the theory, conscious. Maybe C involves a certain kind of behaviorally sophisticated reactivity to inputs (as in the Turing Test), or maybe C involves structured meta-representations of a certain sort, or information sharing in a global workspace, or whatever. Unless you possess a fairly unusual and specific theory, probably the following will be true: Not only the whole animal (alternatively, the whole brain) will meet criterion C. So also will some subparts of the animal and some larger systems to which the animal belongs.

If there are relatively functionally isolated cognitive processes, for example, they will also have inputs and outputs, and integrate information, and maybe have some self-monitoring or higher-order representational tracking -- possibly enough, in at least one subsystem, if the boundaries are drawn just so, to meet criterion C. Arguably too, groups of people organized as companies or nations receive group-level inputs, engage in group-level information processing and self-representation, and act collectively. These groups might also meet criterion C.[1]

Various puzzles, or problems, or at least questions immediately follow, which few mainstream theorists of consciousness have engaged seriously and in detail.[2] First: Are all these subsystems and groups conscious? Maybe so! Maybe meeting C truly is sufficient, and there's a kind of consciousness transpiring at these higher and/or lower levels. How would that consciousness relate to consciousness at the animal level? Is there, for example, a stream of experience in the visual cortex, or in the enteric nervous system (the half billion neurons lining your gut), that is distinct from, or alternatively contributes to, the experience of the animal as a whole?

Second: If we want to attribute consciousness only to the animal (alternatively, the whole brain) and not to its subsystems or to groups, on what grounds do we justify denying consciousness to subsystems or groups? For many theories, this will require adjustment to or at least refinement of criterion C or alternatively the defense of a general "exclusion postulate" or "anti-nesting principle", which specifically forbids nesting levels of consciousness.

Suppose, for example, that you think that, in humans, consciousness occurs in the thalamacortical neural loop. Why there? Maybe because it's a big hub of information connectivity around the brain. Well, the world has lots of hubs of complex information connectivity, both at smaller scales and at larger scales. What makes one scale special? Maybe it has the most connectivity? Sure, that could be. If so, then maybe you're committed to saying that connectivity above some threshold is necessary for consciousness. But then we should probably theorize that threshold. Why is it that amount rather than some other amount? And how should we think about the discontinuity between systems that barely exceed the threshold versus barely fall short?

Or maybe instead of a threshold, it's a comparative matter: Whenever systems nest, whichever has the most connectivity is the conscious system.  But that principle can lead to some odd results. Or maybe it's not really C (connectivity, in this example) alone but C plus such-and-such other features, which groups and subsystems lack. Also fine! But again, let's theorize that. Or maybe groups and subsystems are also conscious -- consciousness happens simultaneously at many levels of organization. Fine, too! Then think through the consequences of that.[3]

My point is not that these approaches won't work or that there's anything wrong with them. My point is that this is a fundamental question about consciousness which is open to a variety of very different views, each of which brings challenges and puzzles -- challenges and puzzles which philosophers and scientists of consciousness, with a few exceptions, have not yet seriously explored.

--------------------------

Notes

[1] For an extended argument that the United States, conceived of as an entity with people as parts, meets most materialist criteria for being a consciousness entity, see my essay here. Philip Pettit also appears to argue for something in this vicinity.

[2] Giulio Tononi is an important exception (e.g., in Oizumi, Albantakis, and Tononi 2014 and Tononi and Koch 2015).

[3] Luke Roelofs explores a panpsychist version of this approach in his recent book Combining Minds, which was the inspiration for this post.

[image source]

Flavors of Group Consciousness: Vanilla, Strawberry, and Chunky Monkey with Extra Nuts

Yesterday, I was rereading Philip Pettit's 2018 article "Consciousness Incorporated". Due to some vocabulary mismatch, I find his exact commitments on group phenomenal consciousness not entirely clear [note 1]. (By "consciousness" or "phenomenal consciousness" I just mean conscious experience, the stream of experience, or "something-it's-like-ness" in a relatively theoretically innocent sense.)

Pettit endorses group consciousness of some flavor. But what flavor? A mild flavor, he hopes: something "sufficient to engage philosophical interest" but not too "challenging and mysterious" (p. 33). In contrast, in my article "If Materialism Is True, the United States Is Probably Conscious", I see myself as defending a radical position that clashes sharply with ordinary common sense. So the question is, can we distinguish among different degrees of ontological commitment in endorsing "group consciousness", with vanilla on one end (palatable to almost everyone) and, on the other end, well, let's call it "Chunky Monkey with Extra Nuts".

Group Consciousness: Vanilla

Sometimes a group of people all, or mostly, share a particular conscious state -- in a weak or innocuous sense of "sharing". Individually, everyone (or almost everyone, or at least enough of the group) is undergoing that type of conscious experience. So if I say that the theater audience was alarmed by the sudden collapse of the lead actor onstage, or if I say that World Cup viewers around the globe saw the amazing goal, and if we assume that the alarm and the seeing are conscious experiences, then in a certain innocuous sense the groups share conscious experiences.

(One complication: The alarm or the seeing might manifest differently in different members of the group, depending on, e.g., their mood and their viewing position. Set this aside for simplicity.)

Here's a depiction:

[as always, click to clarify and enlarge]

"The audience felt alarmed by the actor's sudden collapse": In this vanilla version of group consciousness, that statement only implies that (enough) of the audience experienced, as individuals, a feeling of alarm (conscious state A in the depiction above).

Pettit clearly wants something more flavorful than this.

Group Consciousness: Chunky Monkey with Extra Nuts

A radical view of group consciousness, in contrast, posits the existence of a stream of experience possessed by the group in addition to the streams of experience possessed by each individual. I have argued that the United States might have a distinctive stream of experience over and above the experiences had by individual citizens and residents of the United States. If streams of experience, or centers of subjectivity, are discrete, countable things (they might not be), and the group contains N members, then on the Chunky-Monkey-Extra-Nuts view, there are N+1 discrete streams of conscious experience -- 300,000,000-ish for the individual members of the United States plus another one for the group as a whole.

Furthermore, on a view of this sort, the conscious experiences of the group mind might be very different from the conscious experiences of any individual members of the group. If the United States is a conscious entity, for example, it might consciously enforce an embargo. But what it feels like, subjectively from the inside, to enforce an embargo might be completely opaque to any individual person. (Alternatively, consider a possible human-grade group mind that is composed out of smaller insect-grade individual minds, capable of appreciating Shakespeare in a manner far beyond what any insect could do: my Antarean Antheads case).

Here's a depiction:

It is highly counterintuitive (in current mainstream Anglophone culture) to think that the United States, or any existing groups of people, actually give rise to a discrete, higher-level stream of consciousness at the group level -- a distinct locus of subjectivity. On this view, group-level mental states arise from, and are not merely composed of, the mental states (and other interactions) of the members, so that there are four, not three, distinct occurrences of experience A (three among the individuals and a fourth for the group) as well as the possibility of experiences (B, D, E) that occur in none of the individuals. If you find this a weird and radical view, you are probably understanding it correctly.

Pettit presumably doesn't want to defend this flavor of group consciousness. [Note 2]

Group Consciousness: Strawberry

Can we and Pettit find an intermediate flavor -- more interesting than vanilla but not as wild as extra nuts?

Pettit compares the relation that a group mind (or "agent") has to its members to the relation of that a statue has to the molecules composing it:

As the statue relates to its molecules, so the group agent relates to its members. The group agent is not the same agent as the set of its members, because the set of members is not, as such, an agent at all. But still, the group agent is a set of members -- a suitably organized or networked set -- and qua set it is the same collection as the set of members who make it up. The group agent is distinct from the members under the one aspect but not distinct from them under the other (p. 23).

This physical analogy captures the intended non-radicalness of Pettit's view. It is a little too simple, however, since not everything in the members' minds belongs to the group mind, and since the group can have mental states that none of the members individually possess. This isn't analogous to how we normally think of the molecular composition of statues.

A favorite example of Pettit's is the following: The group has three members. Member A believes P, Q, and not-R. Member B believes P, not-Q, and R. Member C believes not-P, Q, and R. No one believes P-and-Q-and-R. The group decides collectively, however, that P-and-Q-and-R is a view they can stand behind as a group. They might endorse "We believe P&Q&R" -- though not all of them even need to endorse that, as long as there's a procedure by which it comes to constitute the group's view, for example, by being voiced by the leader after a proper consultative process.

We might depict the situation thus:

The conscious experience of the group is in the red box: The group consciously believes P&Q&R. It's not enough for the members to share a conscious state (e.g., A), and no individual believes P&Q&R, but due to structural features of their relationships, the group believes P&Q&R in virtue of the right members accepting that "we believe P&Q&R". (Let's ignore the trickier case in which the group believes P&Q&R without any individual member accepting that the group believes this.)

Now, is this an interestingly intermediate "strawberry" flavor of group consciousness? Maybe! But here's a question: In virtue of what is "P&Q&R" a conscious belief that the group possesses? If P&Q&R is a conscious belief because individual group members consciously endorse P and Q and R and/or P&Q&R in the right kind of coordinated way, then maybe this is a fairly vanilla view after all: Conscious experience is still the province of individual people. What Pettit adds is only a somewhat more complex way of picking out which individual conscious experiences count as the group's shared conscious experience. Group consciousness is just individual consciousness, plus a criterion for attributing some of those states to the group as a whole.

On the other hand, if the social relationships among the group members yield more than that, if the group's conscious experience arises from the interconnections among members so that conscious experiences at the group level aren't just individuals' conscious experiences plus a criterion -- well then maybe we're starting to get into Chunky Monkey territory after all.

Suppose there's something it's like to consciously think, "Ah, P&Q&R, that's right!" On the Chunky Monkey view, this experience could really transpire in the group entity, even if it occurs in no individual member's head. On the strawberry-that's-basically-vanilla view, that's impossible, and to say that the group consciously endorses P&Q&R is only to say something about structural relationships among what individual group members do consciously endorse.

----------------------------------------

Note 1: Pettit prefers "coawareness", which he appears to equate with "access consciousness" in Ned Block's sense. He says that access consciousness implies there being "something it's like" and maybe vice versa (at least for the case of belief). Despite this, he says he is "setting aside" the issue of phenomenal consciousness -- perhaps thinking of "phenomenal consciousness" as a phrase that is more theoretically commissive than I hear it as being (see p. 12-14, 33).

Note 2: In footnote 6, for example, Pettit favorably cites his sometimes-coauthor Christian List's 2018 criticism of my article on USA consciousness.

How to Turn Five Discrete Streams of Consciousness into a Murky Commingling Bog

Here's something you might think you know about consciousness: It's unified and discrete.

When we think about the stream of conscious experience -- the series of conscious thoughts, perceptual experiences, felt emotions, images, etc., that runs through us -- we normally imagine that each person has exactly one such stream. I have my stream, you have yours. Even if we entertain very similar ideas ("the bus is late again!") each of those ideas belongs determinately to each of our streams of experience. We share ideas like we might share a personality trait (each having a full version of each idea or trait), not like we share a brownie (each taking half) or a load (each contributing to the mutual support of a single whole). Our streams of experience may be similar, and we may influence each other, but each stream runs separately without commingling.

Likewise, when we count streams of experience or conscious entities, we stick to whole numbers. It sounds like a joke to say that there are two and a half or 3.72 streams of conscious experience, or conscious entities, here in this room. If you and I are in the room with a snail and an anesthesized patient, there are either two conscious entities in the room with two streams of conscious experience (if neither snails nor people in that type of anesthesized state have conscious experiences), or there are three, or there are four. Even if the anethesized patient is "half-awake", being half-awake (or alternatively, dreaming) is to be fully possessed of a stream of experience -- though maybe a hazy stream with confused ideas. Even if the snail isn't capable of explicit self-representation of itself as a thinker, if there's anything it's like to be a snail, then it has a stream of experience of its own snailish sort, unlike a fern, which (we normally think) has no conscious experiences whatsoever.

I find it hard to imagine how this could be wrong. And yet I think it might be wrong.

To start, let's build a slippery slope. We'll need some science fiction, but nothing too implausible I hope.

At the top of the slope, we have five conscious human beings, or even better (if you'll allow it) five conscious robots. At the bottom of the slope we have a fully merged and unified entity with a single stream of conscious experience. At each step along the way from top to bottom, we integrate the original five entities just a little bit more. If they are humans, we might imagine growing neural connections, one at a time, between their brains, slowly building cross-connections until the final merged brain is as unified as one could wish. If necessary, we could slowly remove and reconfigure neurons during the process so that the final merged brain is exactly like a normal human brain.

Since the human brain is a delicate and bloody thing, it will be much more convenient to do this with robots or AI systems, made of silicon chips or some other digital technology, if we are willing to grant that under some conditions a well-designed robot or AI system could have a genuine stream of consciousness. (As intuitive examples, consider C3P0 from Star Wars or Data from Star Trek.) Such systems could be slowly linked up, with no messy neurosurgery required, and their bodies (if necessary) slowly joined together. On the top of the slope will be five conscious robots, on the bottom one conscious robot.

The tricky bit is in the middle, of course. Either there must be a sudden shift at exactly one point from five streams of experience to one (or four sudden shifts from exactly 5 to 4 to 3 to 2 to 1), as a result of ever so small a change (a single neural connection), or, alternatively, streams of experience must in some cases not be discretely countable.

To help consider which of these two possibilities is the more plausible, let's try some toy examples.

You and me and our friends are discretely different animals with discretely different brains in discretely different skulls, with only one mouth each. So we are used to thinking of the stream of conscious experience like this:

The red circles contain what is in our streams of conscious experience -- sometimes similar (A and A', which you and I share), sometimes different (C belongs only to you), all reportable out of our discrete mouths, and all available for the guidance of consciously chosen actions.

However, it seems to be only a contingent fact about the biology of Earthly animals that we are designed like this. An AI system, or an alien, might be designed more like this:

Imagine here a complex system with a large pool of representations. There are five distinct verbal output centers (mouths), or five distinct loci of conscious action (independent arms), each of which draws on some but not all of the pool of representations. I have included one redundant representation (F) and a pair of contradictory representations (B and -B) to illustrate some of the possible complexity.

In such a case, we might imagine that there are exactly five streams, though they overlap in some important way.

But this is still much simpler than it might be. Now imagine these further complications:

1. There is no fixed number of mouths or arms over time.
2. The region of the representational pool that a mouth or arm can access isn't fixed over time.
3. The region of the representational pool that a mouth or arm can access isn't sharp-boundaried but is instead a probability function, where representations functionally nearer to the mouth or arm are very likely to be accessible for reporting or action, and representations far from the mouth or arm are unlikely to be accessible, with a smooth gradation between.

This picture aims to capture some of the features described:

Think of each color as a functional subsystem. Each color's density outside the oval is that system's likelihood, at any particular time, of being available for speech or action in that direction. Each color's density inside the oval is the likelihood of representations in that region being available to that subsystem for speech or action guidance. With a rainbow of colors, we needn't limit ourselves to a discretely countable number of subsystems. The figure also might fluctuate over time, if the probabilities aren't static.

In at least the fluctuating rainbow case, I submit, countability and discreteness fail. Yet it is a conceivable architecture -- a possible instantiation of an intermediate case along our slippery slope. If such an entity could host consciousness, and if consciousness is closely related to its fluctuating rainbow a structural/functional features, then the entity's stream(s) of conscious experience cannot be effectively represented with whole numbers. (Maybe we could try with multidimensional vectors.)

Is this too wild? Well, it's not inconceivable that octopus consciousness has some features in this direction (if octopi are conscious), given the distribution of their cognition across their arms; or that some overlap occurs in unseparated craniopagus twins joined at the head and brain; or even -- reading Daniel Dennett in a certain way -- that we ourselves are structured not as differently from this as we normally suppose.

----------------------------------

Related:

A Two-Seater Homunculus (Apr 1, 2013);

How to Be a Part of God's Mind (Apr 23, 2014);

If Materialism Is True, the United States Is Probably Conscious (Philosophical Studies, 2015);

Are Garden Snails Conscious? Yes, No, or *Gong* (Sep 20, 2018).

Two Ways of Being a Group Mind: Synchronic vs. Diachronic

Based on last week's post, I am now seeing ads for sunglasses everywhere, as if to say "Welcome, Eric, to the internet hypermind! Did you say SUNGLASSES?!"

Speaking of hyperminds....

I want to distinguish two ways of being a group mind, since I know you care immensely about the cognitive architecture of group minds I'm a dork. My thought is that the philosophical issues of group consciousness and personal identity play out differently in the two types of case.

Synchronic:

Examples: Star Trek's Borg (mostly), Ann Leckie's ancillaries, highly interconnected nations (according to me), Vernor Vinge's tines.

Synchronic group minds are probably the default conception. A bunch of independent, or semi-independent, or independent-enough entities remain in constant or constant-enough communication with each other. Their communication is sufficiently rich or sufficiently well structured that it gives rise to group-level mental states in the whole. In the most interesting case, the individual entities are distinct enough to have rich mental lives but also the group is well enough coordinated that it also, simultaneously, has a distinctive, rich mental life above and beyond that of its members.

Diachronic:

Examples: David Brin's Kiln People, Linda Nagata's ghosts, Benjamin Kinney's forks.

In a diachronic group mind, communication is relatively rare but high bandwidth and transformative. Post-communication, the individuals inherit mental states from the others. In the most interesting case, the inheritance process is very different from just listening credulously to someone's testimony; rather it's more like a direct transfer of memories, plans, and opinions, maybe even values or personality. Imagine "forking" into three versions of yourself, going your separate ways for the day, and then at the end of the day merging back into a single individual, integrating the memories and plans of each, and averaging any general changes in value, opinion, or personality. Tomorrow and every day thereafter, you will fork and merge in the same way.

[Cerberus might not be integrated enough to be a good example of a group mind, but I didn't want to attach another darn picture of the Borg.]

Tradeoffs Between Group-Level and Individual-Level Personhood and Autonomy:

As I have described it here, delay between information transfer episodes is the fundamental difference between these types of group minds: whether the minds are in constant or constant-enough communication, or whether instead they communicate only at long intervals. Obviously, temporal distance admits of degree, but this difference in degree creates structural pressures. If communication is infrequent, its effects have to be radical if it is to give rise to an entity sufficiently integrated to be worth calling a "group mind". If every day group of friends meets to exchange information and plan the next day's activities, in the ordinary way people sometimes do this, I suppose that in some weak sense they have formed a group mind. But they haven't done so in the radical science-fictional sense I'm imagining. For example, if there were five friends who did this, there would still be exactly five persons -- entities with serious rights whose destruction would we worth calling murder. For the emergence of something more metaphysically and morally interesting, the exchange has to be radical enough to challenge the boundaries of personal identity.

Conversely, if communication is constant and its effects are radical, it's not clear that we have a group of individuals in any interesting sense: We might just have a single non-group entity that happens to be spatially scattered (as in my Martian Smartspiders).

In other words, to be a philosophically-interesting group entity there must be some sort of interestingly autonomous mentality both at the individual level and at the group level. Massive transformative communication (as in diachronic merging of memories and values) radically reduces autonomy: If communication is both massively transformative and very frequent, there's no chance for interesting person-like autonomy at the individual level. If communication is neither massively transformative nor very frequent, there's no chance for interesting person-like autonomy at the group level.

Consciousness:

Our intuitive judgments about group-level consciousness are probably pretty crappy (as I've argued here and here). But our general theories about consciousness as they apply to the group level are probably even crappier (as I've argued here and here). At the same time, whether the group as a whole has a stream of conscious experience over and above the consciousness of its individual members seems like a very important question if we're interested in its mentality and whether it deserves moral status as a person. So we're kind of stuck. We'll have to guess.

Plausibly, in the diachronic case there is no stream of consciousness beyond that of the merging individuals. When there's one body at night, there's one stream of consciousness (at most, if it's dreaming). When there are three bodies off doing their thing, there are three streams of consciousness. We might be able to create some problematic boundary cases during the merge, but maybe that's marginal enough to dismiss with a bit of hand waving.

The synchronic case is, I think, more conceptually challenging with respect to consciousness. If we allow that minimally interactive groups do not give rise to group level consciousness and we also allow that a fully informationally integrated but spatially distributed entity does give rise to consciousness, it seems that we can create a slippery slope from one case to the other by adding more integration and communication (for example here). At some point, if there is enough coherent behavior, self-representation, and information exchange at the group level, most standard functionalist views of consciousness (unless they accept an anti-nesting principle) should allow that each individual member of the group would have a stream of experience and also that there would be a further, different stream of experience at the group level. But it's a tricky question how much integration and information exchange, and with what kind of structural properties, is necessary for group-level consciousness to arise.

Personhood:

One interesting issue that arises is the extent to which an individual's beliefs about what counts as "self-interest" and "death" define the boundaries of their personhood. Consider a diachronic case: You are walking back home after your day out and about town, with a wallet full of money and interesting new information about a job opportunity tomorrow, and you are about to merge back together with the two other entities you forked off from this morning. Is this death? Are "you" going to be gone after the merge, your memories absorbed into some entity who is not you (but who you might care about even more than you care about yourself)? In walking back, are you magnanimously sacrificing your life to give your money and information to the entity who will exist tomorrow? Would it be more in your self-interest to run away and blow your wad on something fun for this current body? Or, instead, will it still be "you" tomorrow, post-merge, with that information and that money? To some extent, in unclear cases of this sort, I think it might depend on how you think and feel about it: It's to some extent up to you whether to conceptualize the merging together as death or not.

A parallel issue might arise with synchronic groups, though my hunch is that it would play out differently. Synchronic groups, as I'm imagining them, don't have identity-threatening splits and merges. The individual members of synchronic groups would seem to have the same types of rights that otherwise similar individuals who aren't members of synchronic group minds would have -- rights depending on (for example, but it's not this simple) their capacity to suffer and think and choose as individuals. They might choose, as individuals, to view the group welfare as much more important than their own welfare (as a soldier might choose to die for sake of country); but unless there's some real loss of autonomy or consciousness, this doesn't threaten their status as persons or redefine the boundaries of what counts as death.

Related:

Possible Architectures of Group Minds: Perception (May 4, 2016)

Possible Architectures of Group Minds: Memory (Jun 14, 2016)

Group Minds on Ringworld (Oct 24, 2012)

If Materialism Is True, the United States Is Probably Conscious (academic essay in Philosophical Studies, 2015)

Our Moral Duties to Monsters (Mar 8, 2014)

Choosing to Be That Fellow Back Then: Voluntarism about Personal Identity (Aug 20, 2016).

[image source]

Interview on Crazyism

Julia Galef at Rationally Speaking interviews me about the idea that something that seems "crazy" must be true in ethics and the metaphysics of mind.

Along the way, we discuss, among other things, Solar-System-sized orgasm machines, the possibility that the U.S. literally has conscious experience over and above the experiences of the individual people composing it, and the awesomeness of Jorge Luis Borges.

The interview is about an hour long. Listen during your commute or workout!

The written transcript is also available in PDF and DOCX.

Possible Architectures of Group Minds: Perception

My favorite animal is the human. My favorite planet is Earth. But it's interesting to think, once in a while, about other possible advanced psychologies.

Over the course of a few related posts, I'll consider various possible architectures for superhuman group minds. Such minds regularly appear in science fiction -- e.g., Star Trek's Borg and the starships in Ann Leckie's Ancillary series -- but rarely do these fictions make the architecture entirely clear.

One cool thing about group minds is that they have the potential to be spatially distributed. The Borg can send an away team in a ship. A starship can send the ancillaries of which it is partly composed down to different parts of the planet's surface. We normally think of social groups as having separate minds in separate places, which communicate with each other. But if mentality (instead or also) happens at the group level, then we should probably think of it as a case of a mind with spatially distributed sensory receptors.

(Elsewhere, I've argued that ordinary human social groups might actually be spatially distributed group minds. We'll come back to that in a future post, I hope.)

So how might perception work, in a group mind?

Central Versus Distributed Perceptual Architecture:

For concreteness, suppose that the group mind is constituted by twenty groups of ten humanoids each, distributed across a planet's surface, in contact via relays through an orbiting ship. (This is similar to Leckie's scenario.)

If the architecture is highly centralized, it might work like this: Each humanoid aims its eyes (or other sensory organs) toward a sensory target, communicating its full bandwidth of data back up to the ship for processing by the central cognitive system (call it the "brain"). This central brain synthesizes these data as if it had two hundred pairs of eyes across the planet, using information from each pair to inform its understanding of the input from other pairs. For example if the ten humanoids in Squad B are flying in a sphere around an airplane, each viewing the airplane from a different angle, the central brain forms a fully three-dimensional percept of that airplane from all ten viewing angles at once. The central brain might then direct humanoid B2 to turn its eyes to the left because of some input from B3 that makes that viewpoint especially relevant -- something like how when you hear a surprising sound to your left, you spontaneously turn your eyes that direction, swiftly and naturally coordinating your senses.

Two disadvantages of this architecture are the bandwidth of information flow from the peripheral humanoids to the central brain and the possible delay of response to new information, as messages are sent to the center, processed in light of the full range of information from all sources, and then sent back to the periphery.

A more distributed architecture puts more of the information processing in the humanoid periphery. Each humanoid might process its sensory input as best it can, engaging in further sensory exploration (e.g., eye movements) in light of only its own local inputs, and then communicate summary results to the others. The central brain might do no processing at all but be only a relay point, bouncing all 200 streaming messages from each humanoid to the others with no modification. The ten humanoids around the airplane might then each have a single perspectival percept of the plane, with no integrated all-around percept.

Obviously, a variety of compromises are possible here. Some processing might be peripheral and some might be central. Peripheral sources might send both summary information and also high-bandwidth raw information for central processing. Local sensory exploration might depend partly on information from others in the group of ten, others in other 19 groups of ten, or from the central brain.

At the extreme end of central processing, you arguably have just a single large being with lots of sensory organs. At the extreme end of peripheral processing, you might not want to think about the system as a "group mind" at all. The most interesting group-mind-ish cases have both substantial peripheral processing and substantial control of the periphery either by the center or by other nodes in the periphery, with a wide variety of ways in which this might be done.

Perceptual Integration and Autonomy:

I've already suggested one high integration case: having a single spherical percept of an airplane, arising from ten surrounding points of view upon it. The corresponding low integration case is ten different perspectival percepts, one for each of the viewing humanoids. In the first case, there's single coherent perceptual map that smoothly integrates all the perceptual inputs; in the second case each humanoid has its own distinct map (perhaps influenced by knowledge of the others' maps).

This difference is especially interesting in cases of perceptual conflict. Consider an olfactory case: The ten humanoids in Squad B step into a meadow of uniform-looking flowers. Eight register olfactory input characteristic of roses. Two register olfactory input characteristic of daffodils. What to do?

Central dictatorship: All ten send their information to the central brain. The central brain, based on all of the input, plus its background knowledge and other sorts of information, makes a decision. Maybe it decides roses. Maybe it decides daffodils. Maybe it decides that there's a mix of roses and daffodils. Maybe it decides it is uncertain, and the field is 80% likely to be roses and 20% likely to be daffodils. Whatever. It then communicates this result to each of the humanoids, who adopt it as their own local action-guiding representation of the state of the field. For example, if the central brain says "roses", the two humanoids registering daffodil-like input nonetheless represent the field as roses, with no more ambivalence about it than any of the other humanoids.

Winner-take-all vote: There need be no central dictatorship. Eight humanoids might vote roses versus two voting daffodils. Roses wins, and this result becomes equally the representation of all.

Compromise vote: Eight versus two. The resulting shared representation is either a mix of the two flowers, with roses dominating, or some feeling of uncertainty about whether the field is roses (probably) or instead daffodils (possible but less likely).

Retention of local differences: Alternatively, each individual humanoid might retain its own locally formed opinion or representation even after receiving input from the group. A daffodil-smeller might then have a representation something like this: To me it smells like daffodils, even though I know that the group representation is roses. How this informs that humanoid's future action might vary. On a more autonomous structure, that humanoid might behave like a daffodil smeller (maybe saying, "Ah, it's daffodils, you guys! I'm picking this one to take one back to the daffodil loving Queen of Mars") or it might be more deferential to the group (maybe saying, "I know my own input suggests daffodils, but I give that input no more weight than I would give to the input of any other member of the group").

Finally, no peripheral representation at all: An extremely centralized system might involve no perceptual representations at all in the humanoids, with all behavior issuing directly from the center.

Conceptual Versus Perceptual:

There's an intuitive distinction between knowing something conceptually or abstractly and having a perceptual experience of that thing. This is especially vivid in cases of known illusion. Looking at the Muller-Lyer illusion you know (conceptually) that the two lines minus the tails are the same length, but that's not how you (perceptually) see it.

The conceptual/perceptual distinction can cross-cut most of the architectural possibilities. For example, the minority daffodil smeller might perceptually experience the daffodils but conceptually know that the group judgment is roses. Alternatively, the minority daffodil smeller might conceptually know that her own input is daffodils but perceptually experience roses.

Counting Streams of Experience:

If the group is literally phenomenally conscious at the group level, then there might be 201 streams of experience (one for each humanoid, plus one for the group); or there might be only one stream of experience (for the group); or streams of experience might not be cleanly individuated, with 200 semi-independent streams; or something else besides.

The dictatorship, etc., options can apply to the group-level stream, as well as to the humanoid-level streams, perhaps with different results. For example, the group stream of consciousness might be determined by compromise vote (80% roses), while the humanoid streams of experience retain their local differences (some roses, some daffodils).

To Come:

Similar issues arise for group level memory, goal-setting, inferential reasoning, and behavior. I'll work through some of these in future posts.

I also want to think about the moral status of the group and the individuals, under different architectural setups -- that is, what sorts of rights or respect or consideration we owe to the individuals vs. the group, and how that might vary depending on the set-up.

------------------------------------------------

Related:

Possible Psychology of a Matrioshka Brain (Oct. 9, 2014).

If Materialism Is True, the United States Is Probably Conscious (Philosophical Studies 2015).

[image source, image source]

New Essay in Draft: Is the United States Phenomenally Conscious? Reply to Kammerer

Francois Kammerer has a forthcoming piece in Philosophia responding to my 2015 paper, "If Materialism Is True, the United States Is Phenomenally Conscious". I've drafted a reply to Kammerer.

Abstract:
In Schwitzgebel (2015) I argued that the United States, considered as a concrete entity with people as some or all of its parts, meets all plausible materialistic criteria for consciousness. Kammerer (forthcoming) defends materialism against this seemingly unintuitive conclusion by means of an "anti-nesting principle" according to which group entities cannot be literally phenomenally conscious if they contain phenomenally conscious subparts (such as people) who stand in a certain type of functional relation to the group as a whole. I raise three concerns about Kammerer's view. First, it does not appear to exclude the literal phenomenal consciousness of actually existing groups of people, as one might hope such a principle would do. Second, Kammerer's principle appears to make the literal phenomenal consciousness of a group depend in an unintuitive way on the motivations of individuals within the group. Third, the principle appears to be ad hoc.

Many thanks for reader comments on this earlier post, and especially to Francois Kammerer. Further thoughts, concerns, and comments welcome, either in the comments field below or by email to me.

Kammerer's New Anti-Nesting Principle

Anti-nesting principles, in consciousness studies, are principles according to which one stream of consciousness cannot "nest" inside another. According to such principles, a conscious being cannot have conscious subparts -- at least under certain conditions -- even if it meets all other plausible structural criteria for being a conscious system. Probably the best-known anti-nesting principles are due to Hilary Putnam (1965, p. 434) and Giulio Tononi (2012, p. 297). Putnam's version is presented bare, and almost unmotivated, and has been criticized by Ned Block (1981, p. 74-76). Tononi's version is more clearly motivated within his "Integrated Information Theory" of consciousness, but still (I think) has significant shortcomings.

In this forthcoming paper in Philosophia, Francois Kammerer takes another swing at an anti-nesting principle.

Though relatively neglected, nesting issues are immensely important to consciousness studies. Intuitively or pre-theoretically, it seems very plausible that neither subparts of people nor groups of people are literally phenomenally conscious. (Unless maybe the brain as a whole is the relevant subpart.) If we want to retain this intuitive idea, then either (a.) there must be some structural feature that individuals have, which groups and subparts of individuals do not, which is plausibly necessary for consciousness, or (b.) consciousness must not nest for some other reason even in cases where human groups or subparts would have the structural features otherwise necessary for consciousness.

In "If Materialism Is True, the United States Is Probably Conscious", I argue that human groups do have all the structural features that materialists normally regard as characteristic of conscious systems. A materialist who accepts that claim but wishes nonetheless to deny that groups of people are literally phenomenally conscious might then be attracted to an anti-nesting principle.

Kammerer's principle is a bit complex. Here it is in his own words:

"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."

Short, somewhat simplified version: If the reason a larger entity acts like 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. (I hope that's fair, and not too simple to capture Kammerer's main idea.)

Though Kammerer's anti-nesting principle avoids some of the (apparent) problems with Putnam's and Tononi's principles, and is perhaps the best-developed anti-nesting principle to date, I'm not convinced that we should embrace it.

I'm working on a formal reply (which I'll probably post a link to later), but my main thoughts are three:

First, Kammerer's principle doesn't appear to fulfill the intended(?) role of excluding group-level consciousness among actually existing humans, since it excludes group consciousness in only a limited range of cases.

Second, Kammerer's principle appears to make the existence of consciousness at the group level depend oddly on factors on the individual-person level that might have no influence on group-level functioning (such as whether an individual's thinking of herself as part of the group is emotionally motivating to her, which might vary with her mood even while her participation in the group remains the same, creating "dancing qualia" cases).

Third, it appears to be unmotivated by a general theory that would explain why satisfying or failing to satisfy (A) or (B) would be crucial to the absence or presence of group-level consciousness.

None of these three points would be news to Kammerer, so to make them stick would require more development than I'm going to give them today. But before doing that, I thought I solicit reactions from others -- either to the general issue of anti-nesting principles or to Kammerer's specific principle.

Update Jan. 26, 2016:

I have now drafted a more formal reply essay here.

---------------------------------------------------------

Related posts:

Martian Rabbit Superorganisms, Yeah! (May 4, 2012)

Tononi's Exclusion Postulate Would Make Consciousness (Nearly) Irrelevant (Jul 16, 2014)

The Copernican Sweets of Not Looking Too Closely Inside an Alien's Head (Mar 14, 2014)

Why [X] Should Think the United States Is Conscious (X = Dennett, Dretske, Humphrey) (Winter 2012).

[image source]

An Objection to Group Consciousness Suggested by David Chalmers

For a couple of years now, I have been arguing that if materialism is true the United States probably has a stream of conscious experience over and above the conscious experiences of its citizens and residents. As it happens, very few materialist philosophers have taken the possibility seriously enough to discuss it in writing, so part of my strategy in approaching the question has been to email various prominent materialist philosophers to get a sense of whether they thought the U.S. might literally be phenomenally conscious, and if not why not.

To my surprise, about half of my respondents said they did not rule out the possibility. Two of the more interesting objections came from Fred Dretske (my undergrad advisor, now deceased) and Dan Dennett. I detail their objections and my replies in the essay in draft linked above. Although I didn't target him because he is not a materialist, [update 3:33 pm: Dave points out that I actually did target him, though it wasn't in my main batch] David Chalmers also raised an objection about a year ago in a series of emails. The objection has been niggling at me ever since (Dave's objections often have that feature), and I now address it in my updated draft.

The objection is this: The United States might lack consciousness because the complex cognitive capacities of the United States (e.g., to war and spy on its neighbors, to consume and output goods, to monitor space for threatening asteroids, to assimilate new territories, to represent itself as being in a state of economic expansion, etc.) arise largely in virtue of the complex cognitive capacities of the people composing it and only to a small extent in virtue of the functional relationships between the people composing it. Chalmers has emphasized to me that he isn't committed to this view, but I find it worth considering nonetheless, and others have pressed similar concerns.

This objection is not the objection that no conscious being could have conscious subparts (which I discuss in Section 2 of the essay and also here); nor is it the objection that the United States is the wrong type of thing to have conscious states (which I address in Sections 1 and 4). Rather, it's that what's doing the cognitive-functional heavy lifting in guiding the behavior of the U.S. are processes within people rather than the group-level organization.

To see the pull of this objection, consider an extreme example -- a two-seater homunculus. A two-seater homuculus is a being who behaves outwardly like a single intelligent entity but who instead of having a brain has two small people inside who jointly control the being's behavior, communicating with each other through very fast linguistic exchange. Plausibly, such a being has two streams of conscious experience, one for each homunculus, but no additional group-level stream for the system as a whole (unless the conditions for group-level consciousness are weak indeed). Perhaps the United States is somewhat like a two-seater homunculus?

Chalmers's objection seems to depend on something like the following principle: The complex cognitive capacities of a conscious organism (or at least the capacities in virtue of which the organism is conscious) must arise largely in virtue of the functional relationships between the subsystems composing it rather than in virtue of the capacities of its subsystems. If such a principle is to defeat U.S. consciousness, it must be the case both that

(a.) the United States has no such complex capacities that arise largely in virtue of the functional relationships between people, and

(b.) no conscious organism could have the requisite sort of complex capacities largely in virtue of the capacities of its subsystems.

Contra (a): This claim is difficult to assess, but being a strong, empirical negative existential (the U.S. has not even one such capacity), it seems a risky bet unless we can find solid empirical grounds for it.

Contra (b): This claim is even bolder. Consider a rabbit's ability to swiftly visually detect a snake. This complex cognitive capacity, presumably an important contributor to rabbit visual consciousness, might exist largely in virtue of the functional organization of the rabbit's visual subsystems, with the results of that processing then communicated to the organism as a whole, precipitating further reactions. Indeed turning (b) almost on its head, some models of human consciousness treat subsystem-driven processing as the normal case: The bulk of our cognitive work is done by subsystems, who cooperate by feeding their results into a global workspace or who compete for fame or control. So grant (a) for sake of argument: The relevant cognitive work of the United States is done largely within individual subsystems (people or groups of people) who then communicate their results across the entity as a whole, competing for fame and control via complex patterns of looping feedback. At the very abstract level of description relevant to Chalmers's expressed (but let me re-emphasize, not definitively endorsed) objection, such an organization might not be so different from the actual organization of the human mind. And it is of course much bolder to commit to the further view implied by (b), that no conscious system could possibly be organized in such a subsystem-driven way. It's hard to see what would justify such a claim.

The two-seater homunculus is strikingly different from a rabbit or human system (or even a Betelguesian beehead) because the communication is only between two sub-entities, at a low information rate; but the U.S. is composed of about 300,000,000 sub-entities whose informational exchange is massive, so the case is not similar enough to justify transferring intuitions from the one to the other.

Telepathic Cyborg Rats!

...

It was only a matter of time.

The next step, of course, is aerial transmitters atop our heads which allow direct human brain-to-brain interface without all that slow-paced language business (as envisioned in Churchland 1981).

HT: Nathan Westbrook.

...

SarahAbraham

Through a harmony of dendrites, Sarah came to feel touch upon Abraham’s skin. They were waltzing upon the beach, Ishbak on the immutable piano, and a frond brushed Abraham’s back. Sarah felt the frond as though on her own back. She caressed Abraham’s left shoulder and felt the caress as though upon her own shoulder. Abraham’s right hand was touching the skin under Sarah’s arm, and Sarah felt not only his fingers there in the usual way, but also a new complement: the smoothness of her ribs upon her own right hand. She had been feeling that smoothness for a while, she realized, intermingled with the more familiar touch of Abraham’s left hand upon her right as they danced.

They danced three songs in this manner, then lay together upon the sand. Sarah touched the back of her neck with a twig and saw Abraham scratch his own neck. They made love in a new way.

Sarah gazed upon Abraham as he observed the sky. She called to Ishbak and Midian. Sarah and Abraham lay face down upon blankets while the sons of Keturah touched their backs, and Sarah learned to distinguish Abraham’s sensations from her own. She now had two backs, two bodies.

Through a harmony of retinas, Sarah came to see through Abraham’s eyes. At first, it was a faint tint upon her field of view – her own form, maybe, bent toward the fire pit, as Abraham watched her from the side, her figure like a wraith upon the fire that she more vividly saw. The wraith was jumpy, unpredictable; she could not fully guess Abraham’s saccades as his eyes gathered the scene. Over days, the visions livened and settled. Sarah did not know if she merely learned better to anticipate Abraham’s eye movements or if she instead also gained partial control.

[for the second half of this story, email me at eschwitz at domain: ucr.edu]

A Two-Seater Homunculus

My neighbor Bill seemed like an ordinary fellow until the skiing accident. He hit a tree, his head split open, and out jumped not one but two homunculi. I persuaded them not to flee and sat them down for an interview.

The homunculi reproduce as follows. At night, while a person is sleeping, a female homunculus lays one egg in each of the victim's tear ducts. The eggs hatch and minute worms wiggle into the victim's brain. As the worms grow, they consume the victim's neurons and draw resources from the victim's bloodstream. Although there are some outward changes in the victim's behavior and physiological regulation, these changes are not sufficiently serious to engender suspicion in the victim's friends; the homunculi are careful to support the victim's remaining neural structure, especially by sending out from themselves neural signals similar to what the person would have received had their brain tissue not been consumed. The victim reports no discomfort and suspects nothing amiss.

Each growing homunculus consumes one hemisphere of the brain. Shared neural structures they divide equally between themselves. They communicate by whispering in a language much like English, but twenty times as fast -- much less inter-hemispheric information transfer than in the normal human brain, of course, but as commissurotomy cases show, massive information transfer between the hemispheres is not essential to most normal human behavior. Any apparent deficits are masked by a quick stream of speech between the homunculi, and unlike hemispheric specialization in the human brain, both homunculi receive all inputs and have joint control over all outputs.

Two months later, the person is a two-seater vehicle for brother and sister homunculi. An internal screen of sorts displays the victim's visual input to both of the homunculi; through miniature speakers the homunculi hear the auditory input; tactile input is fed to them by dedicated sensors on their own limbs, etc. They control the victim's limbs and mouth by joint steering mechanisms. Each homunculus is as intelligent as a normal human being, though operating an order of magnitude more quickly due to their more efficient brains (carbon based, like ours, but operating on much different internal principles). When the homunculi disagree about what to do, they quickly negotiate compromises and deferences. When fast reactions are needed, behavior defaults to pre-negotiated compromises and deferences.

But what is it like for Bill? There is no Bill anymore, maybe, though he didn't notice his gradual disappearance. Or maybe there still is Bill, despite the radical change in his neurophysiology? How many streams of experience are there? Two? One for each homunculus but none for Bill? Or one stream only, for the two homunculi, if they are well enough integrated and fast-enough communicating? (How well and quickly integrated would they have to be to share a single stream?) Three streams? One for each homunculus plus one, still, for Bill? Two and a half? One for Bill and one and a half for the two partly-integrated homunculi? Or when counting streams of experience, must we always confine ourselves to whole numbers?

Bill always liked sushi. He never lost that preference, I think. Neither of the homunculi would want to put sushi in their own mouths, though. Bill loved his wife, the Lakers, and finding clever ways to save money on taxes. Bill can still recite Rime of the Ancient Mariner by heart, though neither of the homunculi could do it on their own. When he sees the swing in his backyard, Bill sometimes calls up a fond and vivid memory image of pushing his son on the swing, years ago. When the homunculi consumed his brain, they preserved the information in this memory image between them -- and in many others like it -- and they would draw it up on their visual imagery screen when appropriate. Characteristic remarks would then emerge from Bill, like "Maggie, do you remember how much Ethan loved to ride high on this swing? I can still picture his laughing face!" So spontaneous and natural does it seem to the well-entrenched homunculi to make such remarks, they told me, that they would lose all sense that they are acting. Maybe, indeed, they were no longer acting but really (jointly) became Bill.

I don't know why the homunculi thought I would not be alarmed upon learning all this. Maybe they thought that, as a crazy philosopher who takes literal group consciousness seriously, I'd think of the two-seater homunculus as merely an interesting implementation of good old Bill. But if so, their trust was misplaced. I snatched the homunculi, knocked them unconcious, and shoved them back inside Bill's head. I glued and stapled Bill's skull closed and consulted David Chalmers, my go-to source for bizarre scenarios involving consciousness.

None of what I said was news to Dave. He had been well aware of the homunculus infestation for some time. It had been a closely-held secret, to prevent general panic. But with the help of Christof Koch, a partial cure had just been devised, and news of the infestation was being disseminated where necessary to implement the cure.

The cure works by dissolving homunculi's own skulls and slowly fusing their brains together. Their motor outputs controlling the victim's behavior are slowly replaced by efferent neurons. Simultaneously, the remains of the homuncular bodies are slowly reabsorbed by the victim. At the end of the process, the victim's physiology is very different from what it had been before, but it is a single stream in a single brain, with no homuncular viewing screens or output controls and more or less the victim's original preferences, memories, skills, and behavior patterns.

All of this happened two years ago. Bill never knew the difference and remains happily married, though I had to tell a few white lies about his skiing accident to explain the scars.

Group Minds on Ringworld

In the year 3000, let's suppose, humanity completes its greatest construction project ever: Ringworld, a habitable surface as wide as a planet but spanning an entire planetary orbit -- a ring around a neighboring star with 10,000,000,000,000 square kilometers of living space. A big place!

Earthly nations send colonists. Once on Ringworld, the colonists form independent nations, free of Earthly control. These nations grow and spread. For sociological reasons, let's suppose, Ringworld nations function best with populations near 100,000,000. Once a nation grows much larger than that, it tends either to fission or to stagnate. Now, what type of nation will be well represented on the surface of Ringworld after ten thousand years?

Although it could play out in various ways, the most straightforward answer seems to be: nations that grow fast, then fission, then repeatedly grow again and fission again. Mobility to unpopulated parts of Ringworld, away from competitors, might also be favored. Also, we might expect the most evolutionarily successful nations to have intergenerationally stable developmental resources -- that is, to be such that their fission products tend to develop the same traits that the fissioning parent nations had, i.e., the very traits that made those parents evolutionarily successful. Otherwise, after a few generations, those nations' fission-produced offspring nations will be outcompeted. We might further imagine that the most successful nations employ eugenics: Their governments select a range of DNA strands containing especially desirable traits, which then serve as the genetic basis of the next generation of their citizens; and the governments that do so with the best eye to maximizing their nations' eventual descendant nations, and that do so stably over the generations, are eventually the nations that are best represented on the Ringworld surface.

We might imagine, too, that as the Ringworld surface becomes more crowded, aggression starts to pay. In response, the competing nations develop protective physical borders, grown using nanotechnology and difficult to penetrate without permission. Nations might also strictly limit immigration as contrary to their eugenic plans. If nations are somewhat mobile -- and we might imagine that gravity (or centrifugal inertia) is light and fusion power plentiful -- they might best compete with each other by moving toward opportunities and away from threats, bringing their citizenry and physical defensive borders along with them. Eventually, these defensive borders might gain appendage-like functionality -- e.g., offensive weaponry and the ability to harvest minerals and sources of power. Once this happens, the majority of individual citizens might become largely sedentary, communicating via radio and microwave signals. And once sedentary, size-reduction might be selected for, to reduce the energetic costs of nation-scale movement; and transmission of essential nutrients between citizens might be achieved by purely mechanical means. Furthermore, once free of the demands of individual mobility and individual-level reproduction, citizens might start to specialize ever more narrowly in tasks that serve the reproductive interests of the nation -- or at least the nations whose citizens develop in that direction might in the long run outcompete the nations whose citizens do not.

Over time, as individual citizens shrink and become increasingly specialized, and as the membrane around the nation becomes more functional and more effectively protective of the interior, the overall physical structure of the nation might start to look increasingly like that of what we would call an individual organism that reproduces by fission.

Nations -- at least the evolutionarily most successful ones -- will presumably engage in social intercourse among each other, both cooperatively and competitively. Possibly, some of these nations will evolve so that no single individual citizen is responsible for between-nation communication but rather the communicative efforts arise in a complex way from the citizenry as a whole. If individual citizens become sufficiently small and specialized, and if they learn to communicate with each other non-linguistically (e.g., by direct brain-to-brain stimulation), then it might eventually become the case that no individual citizen can even understand the linguistic communications emitted by her own nation.

A million years passes, during which Earth loses communication with Ringworld. Social pressures on Ringworld favor increasingly sophisticated forms of communication between nations, including the emergence of nation-level art, poetry, song, history, and philosophy -- none of which is comprehensible to the individual citizens of the species of nation that eventually conquers the rest. After these million years, visitors from Earth arrive, and they decide that conscious experience is primarily to be found at the level of nations, not at the level of individual citizens.

Question: At what point in this process did the nations first have nation-level conscious experience?

Might it have been from the very beginning?

New Paper in Draft: If Materialism Is True, the United States Is Probably Conscious

Regular readers will have noticed over the past year a so a variety of posts on group consciousness.  I have finally synthesized my thoughts into a full-length paper draft, which I have posted online here.

Here's the core argument: There seems to be no principled reason to deny entityhood to spatially distributed but informationally integrated beings (such as Martian Smartspiders or Betelgeusian beeheads). The United States can be considered as a concrete, spatially distributed but informationally integrated entity. Considered as such, the United States is at least a candidate for the literal possession of real psychological states, including consciousness. The question, then, is whether it meets plausible materialistic criteria for consciousness. My suggestion is that if those criteria are liberal enough to include both small mammals and alien species that exhibit sophisticated linguistic behavior, then the United States probably does meet those criteria. The United States is massively informationally interconnected and responds in sophisticated, goal-directed ways to its surroundings. Its internal representational states are functionally responsive to its environment and not randomly formed or assigned artificially from outside by the acts of an external user. And the United States exhibits complex linguistic behavior, including issuing self-reports and self-critiques that reveal a highly-developed ability to monitor its evolving internal and external conditions.

I leave it open whether we should (a.) do modus ponens and accept the conclusion, (b.) do modus tollens and deny the antecedent, (c.) see this as a challenge to find a good justification for accepting the antecedent while denying the consequent, or (d.) see this exercise as supporting skepticism about a certain style of metaphysics.

As always, thoughts and comments welcome.

Chinese Room Persona

I just came across this passage in Karl Schroeder's science fiction novel Sun of Suns (from a conversation between high-tech engineer Aubri Mallahan and low-tech Admiral Chaison Fanning):

"Max is a Chinese Room persona, which makes him as real as you or I." She saw his uncomprehending stare, and said, "There are many game-churches where the members of the congregation each take on the role of one component of a theoretical person's nervous system -- I might be the vagus nerve, or some tiny neuron buried in the amygdala. My responsibility during my shift is to tap out my assigned rhythm on a networked finger-drum, depending on what rhythms and sounds are transmitted to me by my neural neighbors, who could be on the other side of the planet for all I--" She saw that his expression hadn't changed. "Anyway, all of the actions of all the congregation make a one-to-one model of a complete nervous system... a human brain, usually, though there are dog and cat churches, and even attempts at constructing trans-human godly beings. The signals all converge and are integrated in an artificial body.  Max's body looks odd to you because his church is a manga church, not a human one, but there's people walking around on the street you'd never know were church-made."

Chaison shook his head. "So this Thrace is... a fake person?"

Aubri looked horrified. "Listen, Admiral, you must never say such a thing! He's real. Of course he's real. And you have to understand, the game-churches are as incredibly important part of our culture. They're an attempt to answer the ultimate questions: what is a person? Where does the soul lie? What is our responsibility to other people? You're not just tapping on a drum, you're helping to give rise to the moment-by-moment consciousness of a real person.... To let down that responsibility could literally be murder.

Although John Searle would likely disagree with Aubri's perspective on the Chinese room, I'm inclined to think that on general materialist principles there's no good reason to regard such details of implementation as sufficient to differentiate beings who really have consciousness from those who don't. We don't want to be neurochauvinists after all, do we?

Why Tononi Should Allow That Conscious Entities Can Have Conscious Parts

On March 23, I argued that eminent theorist of consciousness Giulio Tononi should embrace the view that the United States is conscious -- that is, literally possessed of a stream of phenomenal, subjective experience of its own, above and beyond the experiences of all its citizens and residents considered individually. My argument drew on Tononi's work from 2004 through 2009 arguing that any system in which information is integrated -- that is, virtually any causal system at all! -- is conscious. Tononi's one caveat in those works is that to count as a "system" in the relevant sense, an informational network must not be merely a subsystem within a more tightly integrated larger system. I argued that since the U.S. is a system in Tononi's sense, either it or some more tightly integrated larger system (the whole Earth?) must be conscious by Tononi's lights. While in other posts I had to do some work to show how I thought Daniel Dennett's, Fred Dretske's, and Nicholas Humphrey's views implied group consciousness, Tononi seemed an easy case.

However, my March interpretation of Tononi was out of date.  More recently, (here [in note 9] and here [HT Scott Bakker and Luis Favela]), Tononi has endorsed what I will call an anti-nesting principle: A conscious entity cannot contain another conscious entity as a part. Tononi suggests that whenever one information-integrated system is nested in another, consciousness will exist only in the system with this highest degree of informational integration.

Tononi defends this principle by appeal to Occam's razor, with intuitive support from the apparent absurdity of supposing that a third group consciousness could emerge from two people talking. But it’s unclear why Tononi should put much weight on the intuitive resistance to group consciousness, given his near panpsychism. He thinks photodiodes and OR-gates have a little bit of conscious experience; so why not some such low-level consciousness from the group too? And Occam’s razor is a tricky implement: Although admitting the existence of unnecessary entities seems like a bad idea, what is an “entity” and what is “unnecessary” is often unclear, especially in part-whole cases. Is a hydrogen atom an unnecessary entity once one admits the proton and electron into one’s ontology? What makes it necessary, or not, to admit the existence of consciousness in the first place? It is obscure why the necessity of admitting consciousness in a large system should turn on whether it is also necessary to admit conscious experience in some of its subparts. (Consider my Betelgeusian beeheads, for example.) Tononi’s anti-nesting principle compromises the elegance of his earlier view.

Tononi's anti-nesting principle has some odd consequences. For example, it implies that if an ultra-tiny conscious organism were somehow to become incorporated into your brain, you would suddenly be rendered nonconscious, despite the fact that all your behavior, including self-reports of consciousness, might remain the same. (See Ned Block's "Troubles with Functionalism".) It also seems to imply that if there were a large enough election, with enough different ballot measures, the resulting informational integration would cause the voters, who can be conceptualized as nodes or node-complexes in a larger informational system, to lose consciousness. Perhaps we are already on the verge of this in California? Also, since “greater than” is a yes-or-no property rather than a matter of degree, there ought on Tononi’s view to be an exact point at which higher-level integration causes our human-level consciousness suddenly to vanish. Don’t add that one more vote!

Tononi's anti-nesting principle seems only to swap one set of counterintuitive implications for another, in the process abandoning general, broadly appealing materialist principles – the sort of principles that suggest that beings broadly similar in their behavior, self-reports, functional sophistication, and evolutionary history should not differ radically with respect to the presence or absence of consciousness.

Betelgeusian Beeheads

On the surface of a planet around Betelgeuse lives a species of animals who look like woolly mammoths but who act very much like human beings. I have gazed into my crystal ball and this is what I see: Tomorrow they visit Earth. They watch our television shows, learn our language, and politely ask permission to tour our lands. It turns out they are sanitary, friendly, excellent conversationalists, and well supplied with rare metals for trade, so they are welcomed across the globe.

The Betelgeusians are quirky in a few ways, however. For example, cognitive activity takes them, on average, about ten times longer to execute. This has no overall effect on their intelligence but it does test the patience of conversational partners unaccustomed to the Betelgeusians' slow pace. The Betelgeusians find some tasks cognitively easy that we find cognitively difficult and vice versa. They are baffled by our difficulty with simple logic problems like the Wason Selection Task, but they are impressed by our skill in integrating auditory and visual information.

Over time, Betelgeusians migrate down from their orbiting ship. Patchy accommodations are made for their size and speed, and they start to attend our schools and join our corporations. Later, they start to run for political office, displaying roughly the same range of political virtues and vices. Although Betelgeusians don't reproduce by coitus, they find some forms of physical contact arousing and have broadly human attitudes toward pair-bonding. Marriage equality is achieved. What a model of interplanetary harmony!

Everyone agrees that Betelgeusians are conscious, of course. In fact, the Betelgeusians have a long and impressive academic tradition in philosophy of mind and introspective psychology.

Why do I call them "beeheads", you're surely wondering? Well, it turns out that when you look inside their heads and humps, you find not neurons but rather tens of millions of squirming insects, each a fraction of a millimeter across. Each insect has a complete set of minute sensory organs and a nervous system of its own (compare mymaridae wasps on Earth). The beeheads' behavior arises from complex patterns of interaction among these individually dumb insects. These mammoth creatures are much-evolved descendants of Betelgeusian bees that evolved in symbiosis with a brainless, living hive. (The hive itself began as a simple, non-living physical structure that over evolutionary history slowly incorporated symbiotic organisms -- as Earthly ant hives sometimes do -- which then merged into a larger whole.) The Betelgeusians' giant heads and humps have, altogether, a hundred times as many neuron-like cells as the human brain has; and the insects' interactions are so informationally efficient that neighboring insects can respond differentially to the behavioral or chemical effects of other insects' individual efferent neural impulses. The process, still, is an order of magnitude slower than our own.

Maybe there are little spatial gaps between the bees. Does it matter? Maybe, in the privacy of their homes, the bees sometimes fly apart and back together, exiting and entering through the mouth. Does it matter? Maybe if the exterior body is too severely injured, the bees recruit a new body from nutrient tanks -- and when they fly off to do this, they do so without too much interruption of cognition, able to report thoughts mid-transfer. They reconvene and say, "Oh it's such a free and airy feeling to be without a body! And yet it's a fearful thing too. It's good to feel again the power of limbs and mouth. May this new body last long and well! Shall we dance, then, love?"

If you accept the consciousness of Betelgeusian beeheeds, you have accepted the first step in my argument that the United States is conscious.

Martian Rabbit Superorganisms, Yeah!

Most philosophers of mind (but not all) believe that rabbits have conscious experiences -- that rabbits are not, as it were, mere machines all dark inside but rather that there's "something it's like" to be a rabbit, that they can have sensory experiences, that they can experience pain, that they have (in contemporary jargon) "phenomenology". After all, rabbits are not so different from us, biologically.  Rabbits might lack language and higher forms of abstract and self-reflective cognition, but few philosophers think that such differences between us and them are sufficient to render rabbits nonconscious.

Most philosophers of mind (but not all) likewise believe that if we were visited by a highly intelligent naturally-evolved alien species -- let's call them "Martians" -- that alien species might possess a radically different biology from us and yet still have conscious experience.  Outwardly, let's suppose, Martians look rather like humans; and also they behave rather like humans, despite their independent evolutionary origins.  They visit us, learn English, and soon integrate into schools and corporations, maybe even marriages.  They write philosophical treatises about consciousness and psychological treatises about their emotions and visual experiences.  We will naturally, and it seems rightly, think of such Martians as genuinely conscious.  Inside, though, they have not human-style neurons but rather complicated hydraulics or optical networks or the like.  To think that such beings would necessarily be nonconscious zombies, simply because their biology is different from ours, seems weirdly chauvinistic in a vast universe in which complex systems, intelligently responsive to their environments, can and do presumably evolve myriad ways.

Okay, so how about Martian rabbits?  Martian rabbits would be both biologically and behaviorally very different from you and me.  But it seems hard to justify excluding them from the consciousness club, if we let in both Earthly rabbits and Martian schoolteachers.  Right?

Ready for a weirder case?  Martian Smartspiders, let's suppose, are just as intelligent and linguistically sophisticated as the Martian bipeds we love so well.  In fact, we couldn't distinguish the two in a Turing Test.  But the Smartspiders are morphologically very different from bipeds.  Smartspiders have a central body that contains basic biological functions, but most of their cognitive processing is distributed among their 1000 legs (which evolved from jellyfish-like propulsion and manipulation tentacles).  Information exchange among these thousand legs is fast, since in the Martian ecosystem peripheral nerves operate not by the slowish chemical-electrical processes we use but rather by shooting light through reflective capillaries (fiber optics), saving precious milliseconds of reaction time.  Thus the 1000 distributed centers of cognitive processing can be as quickly and tightly informationally integrated as are different regions of our own brains -- and the ultimate output is just as smart and linguistic as ours.  If there were such Turing-Test-passing Martian Smartspiders, it seems we ought to let them into the bounds of genuinely conscious organisms, if we're letting in the bipedal Martians and the Martian rabbits.

Suppose Martian Smartspiders evolve so that their legs become detachable, while still capable of movement and control by the organism as a whole.  The detachment can work because the nerve signals are light-based: The Martian just needs to replace directly connective fibers with transducers that can propagate the light signal across the gap from the surface of the limb to the portion of the central body where that limb had previously been attached.  One can see how detachable limbs might be advantageous in hunting and in reaching into narrow spaces.  Detaching a leg should have negligible impact on cognition speed as long as there are suitable transducers on the surface of the leg and on the surface of the main body where it normally attaches, since the information will cross the gap at lightspeed.  If we put the Martian Smartspider in a sensory deprivation room and disable its proprioceptive system, it might not even be able to tell if its legs are attached or not.

So here's the Smartspider.  Its thousand limbs venture out, all under the constant distributed control of the Smartspider -- just as much control and integration as if they were attached.  She's still a conscious organism, though spatially distributed, right?

Now imagine the Smartspider gets dumb.  As dumb as a rabbit.  Evolutionary pressures support a general specieswide reduction in intelligence.  Now we have a Notsosmartspider.  Is there any good reason to think it wouldn't be conscious if the rabbit is conscious?

Finally, let's take these thoughts home to Earth.  The most sophisticated ant colonies (e.g., large leaf cutter colonies) are as intelligent and sophisticated in their behavior as rabbits.  If we're going to reject biological chauvinism and continguism (prejudice against discontiguous entities) on behalf of the Martians, why not reject those prejudices on behalf of Earthly superorganisms too?  Maybe the colony as a whole has a distinctive, unified stream of conscious experience, of roughly mammal-level intelligence, above and beyond the consciousness of the individual ants (if individual ants even do have individual consciousness).

There are potentially important differences between the Notsosmartspider and an ant colony.  Individual ants might not be as tightly informationally integrated as are the pieces of the Notsosmartspider as I've imagined it, and the rules governing their interaction might be fairly simple, despite leading to sophisticated emergent behavior.  But should we regard such differences as decisive?  As a thought experiment, imagine those differences first in the brain of a single Martian biped.

And if ant colonies are conscious, might the United States be conscious too?

[Revised May 5]