Is it theoretically possible to create consciousnesses in software that is comparable (or better) to our human capacities? Are we going to eventfully write enough code that the code itself becomes self aware?
If you ask Futurist Ray Kurzweil, the answer is a definitive yes. This is precisely what Google has hired him to do.
Kurzweil happily points out that our biology is itself a software process, one that we can ultimately replicate. (1). He summarizes it this way:
Information defines your personality, your memories, your skills. And it really is information. We ultimately will be able to capture that and actually recreate it. (2)
Thus Kurzwel is of the impression that if we could create a virtual 3-dimensional simulation of a human brain inside a computer, right down to the atomic level, it would operate just as our physical brains do, right down to having consciousness.
Of course, not all agree. Philosopher John Searle, perhaps best known for the famous Chinese Room thought experiment, points out some limitations.
The simulation of neuron firings no more guarantees the power of neurons to cause consciousness than the computer simulation of a rainstorm or five-alarm fire guarantees the casual powers of rain or fire. (3)
The implication is that though we may be able to create software the emulates consciousness, deep down, there's no logical pathway for it to actually share the experience of human consciousness.
But if you read deeper into what Searle has to say, he is saying only that creating an artificial intelligence with software alone is not logical, but developing hardware and software together to accomplish this is certainly feasible. In fact, Searle goes on to point out the following:
The brain is a biological organ, like any other, and consciousness is as much a biological process as digestion or photosynthesis...If we can build an artificial heart that pumps blood, why not an artificial brain that causes consciousness. (3)
Thus Searle's point is that software alone containing AI is science fiction, yet replicating human consciousness in an artificial manner merely requires the right hardware and software combination...and a whole lot of know-how.
The Need for a Working Theory
The problem is that there's many viewpoints, and nothing yet testable. Its all conjecture.
But there's nothing wrong with that. Good science often begins in this manner. Think about all the theories that helped direct scientific discovery in just the last couple of centuries. Per Searle, we began with:
...an atomic theory of matter, a germ theory of disease, a genetic theory of inheritance, a tectonic plate theory of geology, a natural selection theory of evolution, a blood-pumping theory of the heart, and even a contraction theory of the muscles. (3)
Why not begin with a theory of consciousness?
A theory of consciousness needs to explain how subjectivity could emerge from physical elements. How do we go from non thinking molecules, arrange them in a certain manner, and somehow produce a brain capable of first person experiences?
On the outset, this seems inconceivable. Yet Zeno's Paradox of the Arrow provides the perfect metaphor for why exactly we should not dismiss this idea.
The paradox offers you an arrow flying through the air. (4) For it to be in motion, it must change its position from one instant to another. Yet in what individual instant can you observe an arrow moving from where it is to where it is not? Therefore motion must not be possible - yet it is.
There are many explanations to this paradox, but what's important is recognizing the acceptability of emergent properties. In the case of Zeno's Arrow, the property of motion cannot be described by assessing the properties of individual parts (instances).
In everyday circumstances we accept emergent properties constantly.
Liquidity - There is nothing liquid about hydrogen atoms or oxygen atoms, yet in the right mixture and temperature, they form liquid water - at what point do they go from non-liquid to liquid?
Literary Meaning - Of the 26 letters in the English alphabet, none imply anything complex. Yet arrange them with other letters to form words, sentences, and paragraphs, and meaning emerges that can illicit emotion. At what point do letters go from symbols to substance?
Motion Pictures - Take a series of still photos, spin them around on a projector wheel, and the feeling of motion fully emerges. At what point do the images go from conveying stillness to motion?
In none of the above examples do we need a defined point at which any of the final properties fully emerge to appreciate this reality. Like Zeno's Arrow, the explanations of these phenomena are not problematic despite having no definitive points of conversion.
As Richard Feynman so famously said,
A paradox is not a conflict within reality. It is a conflict between reality and your feeling of what reality should be like. (5)
Theory of Consciousness
Once we accept the possibility of consciousness as an emergent property, its seems natural to arrive at the idea that consciousness could be the by-product of a massive number of organized synaptic firings. This would act much in the same way the experience of motion emerges from the flow of a large set of organized still images.
If its possible that the brain integrates all of the experiences from a multitude of different sensory channels, ultimately producing a complex storm of activity out of which consciousness emerges, is there any reason to think we cannot recreate this storm artificially?
Circle of Emergence
It would seem this is the bedrock of Ray Kurzweil's motivation (and Google's). Of course, there is one major sticking point...
Emergent properties materialize only when they are observed by a conscious agent.
Perhaps we should really be asking how anything could become self aware without first being aware of itself. Quite the circle. And yet, somewhere between our birth to now, we each managed to cross this bridge.
1 - http://www.2045.com/news/32079.html
2 - http://www.pbs.org/newshour/bb/business-july-dec12-immortal_07-10/
3 - "The Mystery of Consciousness" by John Searle
4 - http://plato.stanford.edu/entries/paradox-zeno/#ParMot
5 - "Ball Lightning: Paradox of Physics" by Paul Sagan