Multiple Realizability

The multiple realizability thesis says that there is more than one way to create a specific mental experience. An alternative way to state this is to say that widely different physical systems can have the same mental experience. The implied consequence is that it is the function of the physical system that is causal to the mental experience, an idea I will refer to as functionalism. Historically, the concept of multiple realizability facilitated the move from identity theory to functionalism. Functionalism is more abstract than the identity theory and can therefore accommodate different physical entities causing the same mental experience.

Examples used to illustrate the claims made by the multiple realizability thesis differ in the level at which the systems differ. At the most fine grained level one may argue that replacing all the Carbon atoms with Silicone atoms in your brain will not alter the function of your brain and, consequently, your mental experiences with a Silicone brain would be the same as with your current Carbon brain. At a higher organizational level, each neuron could be replaced by an electrical device that performs the same function the neuron performs. These electrical devices would be connected to each other in the same way your neurons are currently connected to each other and therefore perform the same function - including producing a mental world. It is only one step further to replace the electrical devices with other components (for example citizens of China) that perform the same function. This system would, according to the multiple realizability thesis, still have - as a whole - a mental experience identical to the one you have.

In these examples the neural architecture - the connections between and functions of the components that are neurons in your brain - remains unaltered. Each individual neuron in the brain is replaced by a silicon-neuron, an electrical device, or a citizen of China. The replacement for the neuron is either something that actually could perform the function but does not exist (like a silicon-neuron) or something that exists but can not perform the function of a neuron (electrical devices, citizen of China).

There is a second group of examples illustrating the claims of the multiple realizability thesis in which the neural architecture is not preserved. In this kind of examples, which I will call non-isoform, the neurons do not need to be replaced, which leads to the comparison of different brains (for example of different species). You can also replace the neurons in a non-isoform system which will result in claims like the one that an ant colony could have - as a whole -mental experiences. Ants are not neurons and they are not organized like a brain but they are individual components that occur in large numbers and form an entity by interchanging information. Systems like this are called "agent-based". Each ant in a colony, each neuron in a brain, and each molecule in a cell is an agent. If any agent-based, non-isoform systems can have mental experiences, then mental experience must be regarded as an epiphenomenon caused by complexity. Multiple realizability is also discussed in non-agent-based, non-isoform systems, namely in computers. Despite appearance, computers are less similar to our brains than an ant colony. Ant colonies, like brains but unlike computers, are dynamic and they are self-organized. (The idea that the queen "rules" an ant colony reflects the political believes of the scientists describing the organization of insect states in Victorian times and not the reality.) The illusion that computers are similar to our brains arises because we use them for purposes we would usually use our brain for (mainly remembering stuff). Despite this, the idea that computers could have mental experiences is popular. It seems that this would make mental experience an epiphenomenon of processing a large amount of information (note how that is different from complexity).

Philosophers can pick any of these examples to illustrate their position with regard to multiple realizability and functionalism. Multiple realizability arguments were first used to support functionalism. These were usually based on comparing similar, yet different, systems that were believed to produce the same mental experience. Putnam, for example, argues that different animal species can experience pain despite having different brains; therefore, identity theory should be replaced by functionalism. Somewhat ironically, later thought experiments involving multiple realizability were used to argue against functionalism. These employed vastly different systems that could mimick the function of a brain in an attempt to make it seem implausible that such a system would have mental experience. Ned Block used the citizen of China communicating with cell phones as such an example.

My personal intuition is that Putnam's examples are perfectly plausible, whereas Ned Block's example is implausible.

I intend to solve this contradiction by testing if the systems used in the examples could actually logicalle function like a brain. I believe that the main oversight in the field was to ignore the temporal dimension of the brain-like function performed by the systems. To give just one example: The citizens of China can easily be imagined to be organized like neurons in a brain by giving them communication devices and instructions on how to process and pass on the information they receive. Given the reaction time of human beings, they would pass information at speeds

...