Sunday, September 21, 2014

Explanation and illusion

This is a sequel to a post I wrote in  June called "The explanatory reductio." The point there was this: Although an explanation is an attempt to understand an accepted fact, sometimes our inability to provide an explanation becomes a reason for rejecting the 'fact' instead. If you can't explain that beautiful creature in your bed, maybe it's not your bed.  Here I offer the following observation, and provide you with some examples:

Many of the greatest intellectual insights in human history resulted from someone explaining a widely accepted fact, as a grand illusion.

1. The moon illusion

As everyone knows, the moon is larger when it appears on the horizon than when it is high in the night sky. Sometime back in human prehistory ancient people wondered: Why does the moon grow and shrink like that? One evening, while sitting at his fire after a fruitless day of hunting, the famous cave philosopher Og, had an epiphany.

Og say moon not get small. Moon only look  small. Moon like prey, get small when run away.

Og was the first to articulate the appearance/reality distinction, an incredible leap in human understanding. Later astronomers accepted Og's view that the changing size of the moon is illusory, but they rejected his explanation. They believed all celestial bodies move in perfect circles, with the earth at the exact center of their orbits.  Aristotle proposed a new explanation: The atmosphere acts as a lens, magnifiying the moon's appearance. Wrong again, as can be easily shown. Cover the moon with a quarter while it is on the horizon, and cover it again when it is high in the sky. The distance between the coin and your eye is the same. There is still no accepted explanation of this phenomenon, and there probably will not be one until we understand which of several different illusions relating to relative size and distance our brain circuitry is falling for.

2. Planetary retrograde

The powerful idea that heavenly bodies move in perfect spheres around the earth was dogged for centuries by the strange phenomenon of planetary retrograde. Long before Aristotle, astronomers from different cultures had observed that planets would sometimes start moving backwards. Apollonius and, later, Ptolemy explained the retrograde as a real phenomenon. Their theory of epicycles held that the planets did not move in simple circles, but in epicycles like this:

The epicycle is the smaller orbit, and at the bottom of the epicyclic orbit, the planet itself really does move backwards. Genius. And wrong.  One of the great insights embedded in the heliocentric system proposed by Nicholas Copernicus is that retrograde motion is an illusion that occurs when earth's orbit either overtakes, or is overtaken by, the orbits of other planets.

3: Galilean relativity

Aristotle's physics accepted as real a variety of phenomena that turn out to be illusions. One of the most important- the basis of his categorical distinction between the heavens and the earth - is that rest and uniform motion are different states of matter. For Aristotle the natural state of an object is rest; it requires no explanation. For an object to move, and to continue moving, a cause is required. This illusion was finally shattered by Galileo (the most famous early adopter of the Copernican view) who realized that motion is not a property of an object at all; it is just an artifact of the chosen reference frame. Any object that can be described as moving in one reference frame, can be described as motionless in another. Galileo's insight was ultimately codified in Newton's First Law of Motion.

4: Darwin's theory of evolution

What accounts for the design of the universe? As most philosophy students know, William Paley argued that, just as the evident design of a watch suggests the existence of a watchmaker, so the evident design of the universe suggests the existence of a universe maker. Even David Hume, who rejected the designer hypothesis as childish anthropomorphism, admitted that the design of the universe required an explanation- and that he didn't have one. Enter Charles Darwin. In the Origin of Species Darwin provided a theory of the emergence, not of design, but of the illusion of design. The illusion of design could, he proved, result from a process involving competition, reproduction and blind, natural 'selection'. Darwin's insight effectively completed the Copernican revolution, destroying the nearly universal belief that human origins and human capacities are beyond human understanding.

5: Continental drift

What Darwin did to the eternality of species, Alfred Wegener did to the immobility of continents. Until the late 20th century every schoolkid was taught that these gigantic land masses were necessarily immobile. That uncanny fit between the west coast of Africa and the East coast of South America?  Coincidence. The similar fossil distributions on adjacent coastlines? Strategically placed land bridges that have, sadly, vanished without a trace. Wegener's theory of continental drift, which hypothesized an ancient super continent called Pangaea, represented the immobility of the continents as an illusion. They simply moved too slowly for us to notice. Wegener's theory was widely and understandably ridiculed, for he proposed no mechanism. But long after Wegener's death his view was mostly vindicated by the now universally accepted theory of plate tectonics.

These episodes are exemplary, not just because the thinkers were brave or creative enough to challenge orthodoxy, but because they succeeded in developing a model that explains both the illusion and the real underlying phenomenon.  Even most who achieve this are not ultimately successful. Plato agreed with Parmenides that the reality of the physical world is an illusion.  He explained it poetically as a shadow cast by the real world of the Forms. But the shadow was a metaphor drawn from the physical world, and the perfect world of Forms turns out to exist only in our imaginations. There are many other examples of the kind I have given above, historical and contemporary, successes and failures.  Perhaps our dancers will identify some of them.

G. Randolph Mayes
Department of Philosophy
Sacramento State


  1. Hi Randy, thanks for this. Whats your view on the following?: Take some apparent fact (like quantum indeterminacy). It lacks a complete explanation. What is the most appropriate epistemic position to take? How do we know when the absence of a complete explanation is a reason to doubt the 'fact' or a reason to ramp up the search for a better explanation? I suppose the answer to these questions might have something to do with the 'facts' ability to fit well with our other received facts and help us understand other phenomena.

  2. Dan, thanks for the question. I think the answer is that we don't know, and this is what makes inquiry so difficult, but also so fun. I think the classical empiricist would want to answer that we just have to observe the phenomena more carefully to determine whether things are really happening the way we think we are. But they were still very naive about how observation works It turns out the problem with that idea is that perception is theory laden. Which simply means that our explanatory mechanisms are always involved in the activity of observation. So the best, I think, we can do is remain attentive to the possibility that our explanatory mechanisms are being preyed upon, and that what we think we are seeing is in fact an illusion. (Philosophers are used to this exercise, since it's basically just the argument by skeptical hypothesis.)

    Quantum mechanics is a great example because if we have interpreted the facts right, e.g., that how a particle behaves seems to depend on whether it is measured, it looks like we are forced to give up the locality principle and think of entanglement as requiring instantaneous communication over vast distances. Something Einstein thought he had finally done away with. But today we are starting to see more and more physicists sign on to Everett's Many Worlds interpretation, which proposes that we are interpreting the facts incorrectly, and that the effect of measurement is just to inform us which of many different worlds we inhabit. Many worlds? Action at a distance? At this stage of the game it's a matter of either picking your poison, or just watching to see how things develop.

  3. Randy, that's a very nice catalogue of explained illusions. You say that explanation is an attempt to improve our understanding of the world. It's interesting that we also feel our understanding of the world is improved when we notice or have our attention called to a set of similarities like the ones you point out here. When the similarities are established, we find it satisfying to see diverse phenomena as instances exhibiting a common pattern. In everyday life this might be because it helps us to better predict the way things will behave. When it comes to more abstract things, like theories, revolutions in science, the rise and fall of empires, or the way we overcome illusions, I'm not sure why we find it so satisfying. We feel we understand things better. (But perhaps that's an illusion.)
    But here's a question. In your view, is our belief that we have free will (the ability to cause an action that in principle could not be predicted the way we can predict other natural events) an illusion awaiting an explanation, after which it can be added to your list?

  4. Emrys, thanks for the comment. I think you're right that we get an FOU (feeling of understanding) whenever we feel we've detected a pattern of some kind. And I think you're right that it is related to the expectation of predictive success in everyday examples. I don't see why it shouldn't hold for the more abstract things, too. As you indicate, the feeling is often just sustaining an illusion.

    Free will is a great example, and I'd say that the answer to that is yes,though its complicated by an attractive alternative. I think it was for a very long time regarded as an obvious fact that the mind, in particular, our power of deliberation, is not subject to physical law and therefore properly characterized as an original cause of our rational actions. Clearly in scientific contexts this is now widely understood to be an illusion. I think we have a good explanation of the illusion (we wrongly expect introspection to reveal the causal background of our actions), but I think it's safe to say we are far from having a complete neurobiological account of human choice at this point.

    The complication is that with something like free will we also have the option of saying, not that free will is itself an illusion, but that it is a real phenomenon that we simply understood incorrectly. Which, as you know, is the compatibilist gambit. There are lots of examples of this sort of thing, too. One of my favorites is the concept of knowledge. Skeptics would explain the widely accepted fact that humans have knowledge of the world as an illusion. But skeptical arguments depend on assumptions about the nature of knowledge- the most famous being that knowledge requires certainty- that we choose to reject instead. Some words and ideas are just to central to our understanding of ourselves to give up without a fight.

    Of the examples I discuss above, the only one that seems to have this dimension is Darwin's theory of evolution. If Lamarckian evolution had been universally accepted at the time, Darwin might have chosen to attack the theory of evolution (i.e., the inheritance of acquired characteristics) as an illusion that can be explained by his revolutionary idea of descent with modification.

  5. Randy, your list of suggestive examples strikes me as being a solid case for the claim that sometimes our inability to provide an explanation becomes a reason for rejecting the 'fact' instead. But it doesn’t become the only reason. We aren’t going to give up a fact until we have a good story about why it seems to be a fact but isn’t. When it comes to scientifically interesting facts, and not just the fact that I left my car keys today where I always leave them, that story usually requires a competing, new theory of science that explains the relevant phenomena better than our old theory. Thus we make an inference to the best explanation, that is, we tentatively accept the best of the competing explanations that we know about.

  6. Brad, nice point, I agree completely with this. I think it is probably fairly rare that we recognize an accepted fact as an illusion prior to developing the explanatory model that represents it as such. One possible mechanism for this is for someone to point out some deep, previously unrecognized, absurdity. which most people agree exposes the accepted fact as an illusion even though nobody understands how the illusion is created. But I can't think of any historical examples of this. Can you?