Saturday, May 4, 2013

Philosophical Zombies

Home > Philosophy > Metaphysics

In philosophy, a zombie (also called philosophical zombie or p-zombie) is something that is functionally (and in some cases physically) identical to a human, such that it looks and behaves like a human person but lacks consciousness. Few believe that p-zombies exist in the real world, but some philosophers argue that philosophical zombies being conceivable or logically possible is enough to refute the idea that we are purely physical beings.

A functional zombie is one that is functionally identical to human beings but physically different. The idea that the possibility of functional zombies argues against physicalism (the view that the physical world is all there is) is something I kind of addressed in my The Chinese Room and the Soul article. More commonly though, when philosophers use the term “philosophical zombie” they have in mind not merely functionally identical but also physically identical.

The conceivability of philosophical zombies

In philosophy, the term “conceivability” can mean different things. In some cases a concept being conceivable means mere freedom from self-contradiction; in others it means there is no a priori reason to rule it out (though some might believe that freedom from self-contradiction is the only legitimate form of ruling something out a priori). For our purposes, let’s define “conceivability” to mean that the idea isn’t self-contradictory.

Some philosophers argue that philosophical zombies of the physically identical sort (same electrochemical reactions but no consciousness) are conceivable, and that conceivability of such zombies argues against physicalism. Why think p-zombies are conceivable? Well, upon reflection there doesn’t appear to be anything self-contradictory about the idea of a human brain having the same physical and chemical properties but lacking consciousness. Notably, when it comes to describing all the physical and chemical properties of the brain (certain subatomic particles in certain arrangements moving in certain ways etc.) science has no need for the hypothesis that the moving subatomic particles generate consciousness, and upon reflection there doesn’t appear to be any self-contradiction in those moving subatomic particles not generating consciousness.

Some will find such first-blush intuitions insufficient however and think some additional argument is needed. For this I’ll borrow a bit from John R. Searle, the philosopher who invented the Chinese room thought experiment I discussed in my previous article. Searle argued against the idea that running a computer program would create consciousness. Now suppose a critic says this:
Suppose we create a computer that simulates the actual sequence of neuron firings occurring at the synapses of a native Chinese speaker’s brain when he understands stories in Chinese and gives answers to questions about the story. Surely we would have to say that the computer understands then.
Searle thinks not even this would be enough, and in response he described another interesting thought experiment. Suppose we have a man operate a complex series of water pipes and valves. Given the Chinese symbols as input (the series of symbols constitute questions in Chinese), the rulebook tells him which valves to turn off and on (the rulebook being essentially the computer program). Each water connection corresponds to a synapse in the Chinese person’s brain, and at the end of the process the answer pops out of the pipes (we could supplement this with the water pipes being hooked up to some mechanical apparatus that converts varying water pressure into sound). Unless one wants to posit sentient plumbing here, it seems to me Searle is right about this being insufficient to create consciousness.

The general idea however could be applied to microscopic people invading someone’s brain, disabling the connections of the brain, and setting up a complex cell phone network in its place. Each microscopic person corresponds to a neuron that sends and receives signals. A microscopic person might receive a telephone call and after he hears it ringing, subsequently dials up a few other people, with the pattern of calling exactly mimicking the signaling of neurons had the person been thinking about chocolate. If sentient plumbing does not exist in Searle’s thought experiment, it seems that consciousness would not be generated by our microscopic people calling each other’s phones. And a philosophical zombie is much the same thing as our microscopic cell phone network except that instead of tiny people with cell phones we have neurons—and that difference doesn’t seem it would prevent p-zombies from being conceivable. Upon consideration of all this then, it does seem that philosophical zombies are at least conceivable.

From conceivability to possibility

You might be wondering where the conceivability of p-zombies goes in it being an argument against physicalism. Here’s an argument one could use:
  1. If p-zombies are conceivable, then they are metaphysically possible.
  2. If p-zombies are metaphysically possible, then physicalism is false.
  3. P-zombies are conceivable.
  4. Therefore, if p-zombies are conceivable, then physicalism is false (from (1) and (2)).
  5. Therefore, physicalism is false (follows from (3) and (4)).
Where premises (1) and (2) are best understood as material conditionals. In light of the scenario with the microscopic people, p-zombies not only seem conceivable (thus affirming premise (3)), but also metaphysically possible. Why think premise (2) is true? If there is some possible world where the same physical states (neurons firing etc.) are there but consciousness is lacking, this seems to suggest that something “extra” is needed for consciousness to obtain, and that something like the soul is required.