L.S. Weinrobe | firstname.lastname@example.org
What causes subjective experience, and how can we be sure? Those are the questions at the heart of the hard problem of consciousness. As David Chalmers put it in “Facing Up to the Problem of Consciousness”, his agenda-setting 1995 JCS paper:
For any physical process we specify there will be an unanswered question: Why should this process give rise to experience? Given any such process, it is conceptually coherent that it could be instantiated in the absence of experience.
This quote gets to the root of the hard problem and its grounding in the conceivability argument. For Chalmers, facing up to the problem ultimately means accepting that experience is fundamental and irreducible to any process.
This brief essay offers another option. I will describe a hypothetical process that does not presuppose subjectivity, but that necessarily entails its presence. I’ll then consider how this process might apply to the real world.
First a few definitions:
What is a process? The defining characteristic of a process is change. Whenever a process occurs, it produces a change in the state of one or more entities. And, conversely, whenever there is a change, we can conclude that a process has occurred in order to produce that change.
What is a physical process? Physics, as Bertrand Russell observed, tells us only about the relationships between physical objects and the relative states of the properties of those objects. So a physical process in these terms is one which changes the relationship or relative states of physical objects in some way.
All we can conclude from such a process is that the entities involved have objective reality. They are subject to, and can act on each other in accordance with, the laws of physics. This doesn’t tell us anything about the subjective state of these entities or, more to the point, whether or not they have a subjective state, either individually or collectively. In other words, any such process is subject to the hard problem, as Chalmers asserted.
However, if there was a process that could somehow change an object without changing its relationship to any other object, it would not be subject to the hard problem. The argument goes as follows.
All change is relative. That is, every change is a change in relation to something. So if a process were to act on an entity without altering its relationship to anything external to it, then we must conclude that the entity has changed only in relation to itself.
To say that an entity has changed in relation to itself is the same as saying that it has undergone a subjective change. Subjectivity is the defining characteristic of conscious experience. It’s the explanandum that the hard problem challenges us to explain. Therefore, a process that changes an object without changing its relationship to any other objects would entail the presence of experience.
In short, a subjective change implies a subjective state. If we can find a process – even a hypothetical one – and convincingly show that it produces a subjective change, we will have found an exception to the conceivability argument and a potential new solution to the hard problem.
This seems like a paradox. How can an object change without its relationship to the external world being altered? However, rather than treating it as a paradox, let’s treat it simply as a riddle to be solved.
When is a change not a change? When it’s a translation.
Language offers an analogy. Within its grammatical rules we can alter a sentence in many different ways: by changing the tense of the verb; by modifying the subject or object with an adjective; by changing the punctuation at the end of the sentence to a question mark or exclamation point; and so on. All of these change the context and meaning of the sentence. In our analogy, they are like the many different ways that the relationships or relative states of physical entities can change within the laws of physics.
But we can also change the sentence by translating it into a different language. Leaving aside questions of cultural nuance, if the translation is accurate the meaning stays the same.
One of the intriguing ideas in contemporary physics is that there can be two very different, but corresponding, “languages” of physics. The most well-known example is the AdS/CFT correspondence discovered by the physicist Juan Maldacena. The two theories are populated by different types of entities that are governed by different laws, acted on by different forces and even exist within different numbers of dimensions. Yet every object and every interaction that has a description in one physics also has a corresponding description in the other. This has become a powerful tool for physicists since a phenomenon that may be very difficult to analyze in one theory can be much easier in the other.
This idea of dual theories of physics, I think it’s fair to say, is normally understood to have epistemological rather than ontological significance. Whether we describe the world in English or Chinese doesn’t actually change anything about the world. Similarly, the dual theories are generally seen as alternate ways of understanding or referring to the same reality.
However, we’re looking for a process in which objects actually change, though not in the usual ways described by physics. So for our purposes, let’s borrow the idea of dual physics from holographic models like AdS/CFT, but alter it in a few significant ways. Instead of parallel descriptions of reality, let’s suppose that there are actually two corresponding – and mutually exclusive – physical forms in which reality can express itself, like a bilingual person who can express herself in one language or another depending on where she is and who she’s talking to.
Specifically, the hypothetical scenario I will look at is one where:
- There are two distinct but corresponding modes of physical expression that equate to dual forms of physics. We’ll call them physics A and B.
- There are two neighboring regions or domains which we’ll call region 1 and region 2. Unlike holographic models where the boundary and bulk descriptions mirror each other, regions 1 and 2 are independent in the sense that at any given moment different entities are present and different activity is occurring in the two adjoining regions.
- Region 1 is governed by physics A. That is, all physical entities in region 1 express themselves in their A form. Similarly, region 2 is governed by physics B, and any entities in region 2 express themselves in their B form.
- In our scenario, objects can cross from one region to the other. When they do, they change from expressing themselves in their A form to expressing themselves in their B form, or vice versa.
The next step is to show that this hypothetical process of translation necessarily entails subjectivity, before suggesting how it might apply in reality. As noted above, the key is to prove that the entities that change their form of physical expression through this process do not also change in their relationships to other objects.
First, let’s establish how we would know that a change is an objective change. We would know this if it is a change that is, in principle, detectable from a perspective external to the object or objects that undergo the change. “In principle” to distinguish this from a situation where the inability to detect the change is due to some practical or technical limitations. So even if the change could only be detected at the tiniest, most fundamental scale, that would still be a change that is detectable in principle. And “detectable” as opposed to “observable” since there are changes that are posited in physics – the evolution and the collapse of a wave function in quantum mechanics are good examples – that cannot be observed but that still have a detectable effect.
So the test we’ll apply is this: If a change is not detectable under any conceivable circumstances, then we’ll conclude that it’s not an objective change. And, as argued above, a change that is not objective must be subjective.
In order to test this out, let’s try a thought experiment.
Imagine a tennis court. As in the scenario outlined above, one side of the court is governed by physics A and the other side is governed by a corresponding physics B, with the net as the boundary. Alice stands ready to serve from the A side of the court while on the B side Bob is ready to return Alice’s serve. Standing behind the baseline on the A side is Andy, the line judge. Andy takes his job very seriously, and is equipped with extraordinarily powerful instruments that can detect the location and every other property of the ball with incredible precision. If even the most miniscule part of the ball were to land over the line, he’ll know.
If Andy or Alice could look across the court and see the B side as it really is, it would look impossibly alien to them, and Bob would have a similar reaction if he could see Alice and Andy in their true form. However, because all the information about the B side that reaches Andy – in the form of light, sound, etc. – has had to cross the net to reach him, those signals themselves have been translated from their B form to their A form. In other words, when the subjects in our experiment look across the net, they see nothing unusual, only a continuation of their familiar world.
Alice then serves the ball and Andy detects the change in the ball’s location as it sails over the net, but because he sees everything on the B side of the court translated back into the familiar A forms, he has no way to detect that the ball has also changed from its A form to its B form in crossing the net.
To make it more clear that the transformation of the ball from its A form to its B form is impossible to detect, let’s alter the experiment to eliminate even the change of location as the ball crosses the boundary. We’ll adopt a plot trick from the movies, and imagine that the whole scene is frozen, with a few exceptions.
Alice, Bob, the ball, all the fans in the stands, and everything else in the scene down to the level of the smallest microphysical entities are frozen in place, with the ball stuck in the air just short of the net on the A side. However, Andy is not affected, and can still use his instruments to send out signals and receive the returning signals in order to precisely measure the properties of the ball and its relation to the environment. The other thing that is not impacted by the freeze is the boundary between the A and B regions of the court. Rather than being fixed at the net, the boundary begins shifting towards the A side.
While the ball is still on Andy’s side of the boundary and in its A form, Andy takes detailed measurements of the scene, capturing the location and every other physical property of the ball down to the smallest level of detail and recording how it relates to every other object on or around the court. He then repeats these measurements after the boundary has overtaken the ball, and the ball has been translated to its B form.
However, since the signals coming from the other side of the boundary are translated back into their A form as they cross the boundary, when Andy compares the two sets of measurements he finds no difference whatsoever. Even though the ball has changed from its A form to its B form, its relationship to the external world has not changed in any detectable way. We can therefore conclude that a subjective change has occurred, and that – in some form and at some level – this process of translation was accompanied by a subjective state.
But what could this hypothetical scenario have to do with the real world? The answer goes back to our earlier question.
“When is a change not a change?” also captures a central and longstanding dilemma in the philosophy of time. In choosing between the eternalist and presentist views of time, there are good reasons to accept the eternalist, or block universe, view. One notable reason is Einstein’s special theory of relativity, and specifically the relativity of simultaneity. But if we accept block spacetime, how do we then explain our experience of time as a flowing present?
Attempts to claim that the flowing present is an illusion are not very convincing, any more than attempts to argue that consciousness itself is illusory. However, “moving spotlight” views that try to reconcile block time with a flowing present lead to an apparent paradox. The notion of events being “illuminated” implies that they undergo some sort of temporary change in order to become present. But a basic premise of block spacetime is that all of the events that populate the block are fixed in their relationship to each other, so we’re faced with a familiar question: How can an event change without altering its relationship to other events?
We can resolve this by applying our tennis court scenario to spacetime. In this version, rather than two sides of an imaginary tennis court, the dual domains of physical expression would map to adjacent past and future regions of the four-dimensional block. In other words, the past and future are governed by distinct but corresponding physics, and a given event will be expressed in one or the other form depending on whether it is located in the future or the past.
The other condition I need to add in order to complete the scenario is that the block is static except for the border between the two regions, which steadily moves in the direction of the future. Let’s call this a “shifting boundary” variation of the block view of time. When the boundary overtakes a future event, that event is translated to its corresponding past form, and so on for the next event, and then the one after that as the boundary continues to advance.
As we showed in the tennis court thought experiment, even as events are translated through this process the relationships between events would remain the same, resulting in a series of subjective changes and therefore subjective states. This gives us a mechanism to account both for how experience is generated, and also why that experience is bound to a flowing present.
At the beginning of this essay I acknowledged the challenge presented by the hard problem and the conceivability argument. I noted that the solution Chalmers proposes is to accept that consciousness is fundamental. I then offered an alternative by outlining a process that would entail a subjective state. My hope is that this alternate solution will be of interest to others who may be able to translate these analogies and suggestions into a more detailed and rigorous theory.
But that would still leave two potential solutions on the table. How do we choose between them?
Let’s assume that the block view of spacetime is correct, in the sense that all events are equally real. With that assumption, I think we have a reason to prefer the consciousness-in-translation model over non-reductionist views that treat subjective experience as fundamental.
The reason is this: If all events in time are equally real and subjectivity is a fundamental feature of reality, then we would expect the experience of being present to be equally associated with events throughout the block regardless of their temporal location. In that case, we would expect all the events of our own conscious life to be present equally and at once, either as an integrated whole or perhaps as a number of distinct and coexisting subjective states associated with various moments or temporal segments of our experience. Seen in that light, the fact that our experience is inextricably tied to a flowing present offers evidence for the solution outlined in this essay.
Notes Chalmers, D.J. (1995) Facing up to the Problem of Consciousness, Journal of Consciousness Studies 2(3): 200-19. http://consc.net/papers/facing.html  Russell, B. (1927) The Analysis of Matter. Kegan Paul.  Maldacena, J. (2005) The Illusion of Gravity, Scientific American 293 (5), pp. 56-63. https://www.sns.ias.edu/ckfinder/userfiles/files/sciam-maldacena-3a.pdf  Note that this would be the case even if Andy himself was translated along with the ball. Let’s say that Andy is on the sideline parallel to the ball rather than on the baseline behind Alice. He’s completed his first set of measurements already, then the boundary overtakes him and the ball at the same time. This would cause not only Andy’s physical body and his instruments to be translated to their B forms, but also the data he has recorded about the ball and the environment along with his memories of everything that happened previously. Life in the B domain would be his new normal, without any evidence that things had ever been otherwise. So again when he completes his second set of measurements and compares them to the first, there would be no difference between the two.  To be clear, the claim is that this process would entail a subjective state, but not that the presence of specific qualia can be deduced from the translation of particular physical states. In other words, this is meant to explain why we experience, not what we experience. For the latter, we can turn to panqualityism, the idea that the properties we experience are a fundamental part of reality independently of our experience of them. In his essay “Panpsychism and Panprotopsychism” (The Amherst Lecture in Philosophy 8 (2013): 1–35. http://www.amherstlecture.org/chalmers2013/ ), Chalmers considers panqualityism as a promising form of panprotopsychism, but makes the case that it ultimately falls to the conceivability argument since “…no instantiations of qualities ever necessitate awareness of qualities.” In this light, the translation model supplements panqualityism by identifying a context in which awareness of qualities would be necessary.  Of course I don’t mean to imply that a tennis ball is actually a unitary subject of experience. As to the question of what constitutes a subject of experience, I think that would depend on how the physical entities are organized when they’re translated. To return to the analogy with language, a random group of letters would be translated differently than those same letters rearranged into a word, while the same word may be translated differently on its own or in a sentence where context may alter the word’s meaning. To really apply this analogy would require a detailed description of the corresponding systems of physical expression and how they relate to each other, but the general idea is that a collection of entities can be translated either as a unitary whole or as discrete constituents depending on how they are arranged, and that this determines whether the translation produces a number of subjects with relatively simple phenomenal states or a unified subject with a more complex experience.  Petkov, V. (2005) Is There an Alternative to the Block Universe View?, Dieks, D. (ed.), The Ontology of Spacetime (Elsevier, Amsterdam 2006); “Philosophy and Foundations of Physics” Series, pp. 207–228. http://philsci-archive.pitt.edu/9470/