One way to understand what that means is to consider how the Universe’s ability to rearrange matter/energy into new patterns has gone through paradigm shifts in the past. Any alien intelligences observing our solar system would find examples of patterns that could have come about by chance; matter organised by the driving forces of gravity or geological activity; shaped by the sculpting power of weathering. But, on one planet-Earth- they would encounter numerous examples of patterns too complex to have come about by random chance. On our world, things are divided up into that which looks designed (such as a fish or submarine) and that which does not (such as a cloud or a sand dune). We say something looks designed if its parts are assembled in ways that are statistically improbable in a functional direction.
We also know that ‘things that look designed’ should be divided into ‘things that look designed, and are’ (such as cameras and computers) and ‘things that look designed, but are not’ (eyes and birds). Any life form, indeed, any functional part of a life form, (such as a heart or a liver cell) is a structure whose complexity is orders of magnitude too improbable to have come about by chance- but only if it is assumed that all the luck has to come in one fell swoop. In 1859, Charles Darwin introduced the brilliant idea that when cascades of small chance steps accumulate, you can reach prodigious heights of adaptive complexity.
Through cumulative evolution by natural selection, the universe gatherered together various techniques for putting together increasingly complex patterns with an uncanny illusion of design. The important point to bare in mind, is that the ‘creative toolkit’ of the preceeding era, a time before cumulative selection somehow emerged from random chance, was simply incapable of putting together the complex patterns of matter/energy that we call ‘life’. Gravity may be able to pull on solar nebulae and organise it into planets, but it cannot pull matter together so that it forms a functional system like a heart. Wind erosion can sculpt rock into eye-catching patterns, but it is statistically impossible that it would sculpt rock into something as complex as a tree. Natural selection represents a power to rearrange the building blocks of matter into new patterns that is above and beyond the forces from which it emerged. It does, however, have recognisable limitations. For one thing, the appearance of design in life is, as I said, an illusion. Natural selection has no foresight. It cannot conceptualize anything and then set about realizing its goal. There is a certain amount of randomness in its methodology. Patterns we call life can self-replicate, but they do not copy perfectly. Errors in the replication crop up, leading to subtle differences. Differences that just happen to be disadvantageous are more likely to be wiped out by various environmental factors before they can self-replicate. Differences that happen to give an advantage stand more chance of being imperfectly copied. The way the environment ‘selects’ between accidental ‘bad luck’ and ‘good fortune’ is definitely NOT random. But evolution does not ‘know’ what a good design, or what a bad design, is. It doesn’t ‘know’ anything. It just blindly presents possible solutions to a problem and the environment selects the good from the bad. Nor, for that matter, does the environment ‘know’- at least, not in the sense that a trained architect ‘knows’ that a building she is designing will be structurally sound or not. It is just that, on at least one planet, various factors have come together that, over time, ended up producing a powerful illusion of design and an unmistakable display of complexity.
Another limitation of natural selection, is its inability to comprehensively redesign anything it has built. It can only modify that which it has already made. However, it did assemble matter into a particular complex functional pattern that we call ‘human being’. Thanks to the fact that humans have brains large and organized enough to be able to imagine that which does not exist, the communicative ability to transmit ideas to fellow humans, and the manual dexterity to recombine matter into new patterns, the limitations of natural selection were overcome. Complex functional patterns arose- such as microprocessors- that are not an illusion of design but WERE designed. We can imagine complex functional patterns that do not physically exist and then work out how they might be put together. ‘How’ does not necessarily come to us all at once- all modern technology is the result of generations of accumulated knowledge and the lessons learned from trial and error. But the key thing that differentiates technology from natural selection is that we can think ahead, see a goal and stumble our way toward it. Morover, we can, if needs be, perform complete redesigns. At the moment, the heart of a computer is the microprocessor, but we are drawing up various plans for a whole new paradigm of information processing when the current generation reaches foreseeable limits.
Admittedly, technology does seem like a step backward from natural selection, in that the products of nature are often of a complexity above and beyond that of their technological counterparts. Prosthetic limbs, for example, come nowhere close to matching the ability of natural limbs. But, you have to remember that natural selection has been blindly throwing together possible solutions for billions of years. On the other hand, humans went from stone-age tools to manufacturing all the designs of the modern age in 100,000-2 million years. That seems a long time when judged on human timescales, but in terms of natural selection it is unimaginably rapid. If the fruits of 2 million years-worth of accumulated knowledge were compared to whatever natural selection evolved over a comparable period, the former would almost certainly have the edge in terms of functional elegance and complexity.
So, in our Universe we can identify three paradigm shifts in the power to produce patterns of matter/energy, each ultimately emerging from the products of the preceeding paradigm. That is not to say that a technological species necessarily results from natural selection. It is widely believed by the scientific community that natural selection has no direction and no modern species was more likely to have evolved than any other. But nevertheless, I do believe it would make sense that a tour of the Universe would involve dividing patterns of matter and energy into that which came about through random chance and simple actions; that which evolved a powerful illusion of design through natural selction and that which was intelligently put together to fit a purpose, via technology. It does make sense to describe technology as a new chapter in the story of pattern-building, as distinct from natural selection as natural selection is from random chance.
Right, now finally I get to the point of what the Technological Singularity is. It is a fourth paradigm shift in the ability to create functional patterns. I don’t just mean physical patterns but conceptual ones too. Mental patterns that we label ’creativity’, ’ideas’, ’inspiration’, ’intuition’. Imagine an abstract space in which every kind of mental and physical pattern sits, waiting to be discovered. Some are simple enough for it to be possible that random chance may stumble across them. Others are too statistically improbable and require cumulative selection. Others require foresight if they are to see the light of day. But, now imagine that there exists patterns of a complexity that puts them beyond the grasp of the natural intelligence of human beings. Ideas beyond our ability to imagine, technologies whose complexity is too great. But it is believed by some that certain technologies like biotechnology, information technology and nanotechnology may be combined in such a way that we overcome the limits of human intelligence. Although evolution has clearly put together marvellously complex designs (and the human brain is the most complex design of all) it very rarely produces a design that is of the highest possible functionality. It just doesn’t need to do that. A design of life does not have to be perfect, it only needs to be barely good enough to have a chance of reproducing. We should therefore not be lulled into thinking that the fabulous power of the human brain really is the most capable information processor that can be physically built. Vernor Vinge identified four possible ways in which the fourth paradigm shift aka Singularity might come about:
‘There may be developed computers that are “awake” and superhumanly intelligent’.
‘Large computer networks (and their associated users) may “wake up” as a superhumanly intelligent entity’.
‘Computer/human interfaces may become so intimate that users may reasonably be considered superhumanly intelligent’.
‘Biological science may provide means to improve natural human intellect’.
The first rule of Fight Club is: You don’t talk about Fight Club. And the first rule of the Technological Singularity is: You can’t talk about the Technological Singularity. At least, you can’t talk about it beyond vague terms concerning ‘unknowability’. The fundamental problem concerns effectively describing how it feels to be smarter than a human, and what such a mind might be capable of imagining. Not that this prevents some people from having a go.
Damien Broderick compiles several such attempts in his book, ‘The Spike’. The last chapter deals with what I call ‘technology-driven cosmology’. That is, the suggestion that our future progeny may develop technology of such power, and extend their reach so far, that the evolution of the universe comes under intelligent control. You can see the logic at work here. Take the largest functional pattern you can conceive of (and they don’t come much bigger than a Universe), and make it the plaything of weakly godlike intelligences. After a brief introduction to String theory, a run-of-the-mill standard (ie not technologically-driven) grand Unified Theory that describes our Universe as ‘a mere shadow cast by a richer realm made up of ten dimensions’, we move on to:
Robert Bradbury’s concepts of ‘mega scale super intelligent thought machines’, built from the reprocessed matter of entire planetary systems (a feasible task, so Bradbury assures us, given replicating robotic factories) that ‘consume the entire output of stars’ in order to run enough calculations to simulate populated worlds equal to 10 billion souls for every star in a thousand galaxies.
Jonathon Burns’ idea that ‘maybe the quark-gluon plasma (a state that the universe is believed to have been in, very early in its life) is riddled with quantized chromo dynamic flux tubes in bunches. Assymetry. Structure. Gates and switches’. In other words, once-upon-a-time the whole Universe self-organized into an almighty computer. Hey, maybe your reality is a simulation emerging in the vast amounts of calculations going on somewhere in this computer at the beginning of time? Or, alternatively…
Frank Tipler’s ‘Omega Point’ theory: Our technological descendents collapse the Universe and derive useful computations from the deformation of spacetime. At the final split second before it collapses back to a singularity, an infinite number of computations can be performed which are used to ‘fetch us back to life at the end of time, reconstructing each one of us, in a virtual Universe inside its own mind’ (this last point is proved, Tippler boldly declares, by Game Theory and microeconomic analysis).
Well, there you go. Lots to think about when it comes to the fourth paradigm, huh?