It was a warm evening in May, a little overcast, nothing dramatic. So far so good. I was walking on a disused road in the East End of London over the River Lea, noticing that the area was called the `Greenway' project, a small-scale nature reserve. As I crossed the Lea, with old concrete pill-boxes to my left displaying the occasional desultory graffiti I could see upriver (upcanal actually) towards the Big Breakfast studios, a hint of muted sunset reflected in the water beyond the lock separating tidal from non-tidal regions of the cut, and all of a sudden the song of robin and blackbird impinged on my consciousness. Robin: silver-threaded contralto with rapid and continuous melodic variation; blackbird: a measured improbable baritone traversing a melodic line that only an inebriate could follow. I spotted the male blackbird on a tree and as I approached it I noticed the accurate yellow on its beak and eyes. Turning down to the canal the vegetation was lush, having gained a spurt from the recent warm, moist weather. A heron was perched on the other side of the water, but it looked wrong. As I came to the water's edge it suddenly turned, and I realised that I had been looking at its back it had been staring up a covered culvert in the hope of an evening meal. It did not look away as I turned under the bridge I had just walked over, and I noticed the plaque commemorating the Victorian sewage engineer W.F.Bazelgette as responsible for the 'Northern Outfall Sewer'. Nice touch. Past the bridge I was approaching the lock when my heart missed a beat. I don't often get this premonition, but it is nearly always right. Ahead, in the wall by the canal, there was an old door, but, this time, it was off its hinges. Acting as calmly as I could I slowed my pace a little and turned only my eyes as I drew level with it. And saw nothing. Absolutely nothing. One cannot describe it: it is neither dark nor light, nor full of detail nor blank. Two seconds later my vision lurched and through the broken door was revealed the old house standing there, the house I had only previously seen from the bridge. It was enough. I jacked out, consulted briefly with the other examiners and jotted down our assessment for candidate, code number: G0D03761. Later that week I explained to her that the mark, one short of the 75% required for a distinction, was a tough mark but reached by virtue of the demanding brief she had set herself. `The birds were fabulous,' I told her, `And the social realism was great, particularly the contrast between the well-meaning eco-socialists of the 'Greenway', and the cheerful petty vandals. But you forgot that they had kicked in the door.'

No, this is not an extract from a forthcoming William Gibson novel, but a way of introducing my discipline: computer graphics. I teach artists the use of 3D computer graphics systems, which will one day be used to deliver virtual reality environments like the one in the above scenario. Or will they? I don't regard this as a trivial question, because much depends on it. In the recent writings of physicists Frank Tipler and David Deutsch an assumption is made that humanity will have the option to live in virtual reality as a genuine, or even the only, alternative to 'real' reality. From this assumption some pretty serious implications for humanity are drawn. I am both encouraged and discouraged by this use of a predicted technology upon which to construct a metaphysics. I am encouraged because the whole contemporary phenomenon of writings by scientists on philosophical issues is an intellectual breath of fresh air. I am doubly encouraged because there is an implicit recognition of technology (as recognised by the Victorians in honouring Mr Bazalgette of the Northern Outfall Sewer). But I am discouraged because of the cavalier way in which these two scientists have treated the technology.

The technology matters because technology has always mattered, far more than the history of science gives credit. Newton honoured his predecessors by saying that if he could see so far it was because he stood on the shoulders of giants. But what did he see with? The telescope. Science was born through the observation of the planets by means of the first and most significant piece of research technology ever invented. Earth-bound, science could not arise, because of the difficulty of separating out cause and effect, shown for example in the Greek misconception of force as that which produced velocity. Through the telescope measurements could be made of bodies whose motions, although complex, were affected by simple and clear-cut factors. The success of the gravitational theory in resolving the question of planetary motion gave science the confidence to apply its method to terrestrial problems. In another critical example for modern physics the optical instruments of Michelson and Morely were to establish the invariance of the speed of light, leading to Einstein's theory of relativity. And in the final triumph for the telescope, it was a modern descendant of this humble instrument that, in the hands of Eddington in May 1919, provided experimental proof of a key prediction of Einstein's relativity: the curvature of space near massive bodies.

Technology is no mere 'application' of science, it is as intimately related as the chicken and the egg. After Darwin we can confidently assert of course that the egg came first (its parents were proto-chickens), and I am tempted to say by analogy that the technology is the egg, its parents being merely proto-science. The Greeks could not establish a proper empirical science for two reasons: they were contemptuous of 'research' (both Heraclitus and Plato chide Pythagoras for attempting to understand the world by measuring it), and they had no technology to do it with, even if they had been motivated. The great Greek contribution was of course geometry, and it is of interest to note that Newton's inverse square law of gravitation derives from the geometry of a sphere (as does the inverse-square law in other situations). Anything spreading out from a point in three-dimensional space can be pictured passing through the surface of an imaginary sphere, and diminishes in inverse proportion to the area of the sphere, which is in proportion to the square of its radius. Hence a Greek of phenomenal imagination and geometrical skill could have anticipated the inverse-square law, without ever using a telescope. It is a sobering fact however that the greatest philosophers almost never deduced a physical law from axioms they chose to work with, and very often 'deduced' completely incorrect physical laws, as Descartes did regarding mechanics. It seems that careful observation and measurement are the only basis for science.

Which brings me to the central theme of this essay: the 'new' metaphysics, and how it is related to science and technology. Ever since Aristotle's metaphysics, we have had an instinct to take the science of our time and construct philosophical systems from it. Saint Augustine abandoned nine years of commitment to the Manichean sect (an early rival to Christianity) because a leading Manichean bishop called Faustus could not answer his questions on astronomy. By an irony of fate Augustine helped create the Catholic system that persecuted Galileo for progress in that very question. However, by the time that Newtonian science was established it was intellectually profound enough to spawn a metaphysics of its own, one we could call a reductionist metaphysics. Laplace was the first to find that God was not needed in this mechanical universe, though a Deist metaphysics survived for a while which relegated God to setting the mechanism in motion and then retiring. In the middle of the nineteenth century it would have been safe to assume that physicists would adopt a reductionist metaphysics, while biologists were required to call on more Platonic ideas about living organisms, a vitalist point of view. They could not have anticipated that the twin revolutions of Darwinism and the new physics would reverse the positions, leaving physicists prone to a more fluid, paradoxical view of the world and the biologists as the arch-reductionists.

Hence we can say that the metaphysics of the late 20th century, an outpouring from contemporary scientists, is broadly divided into two: a 'new' reductionism found mainly amongst the Darwinists; and a quite different type of metaphysics with its roots in the 'new' physics. We could call this second type an 'anthropic' metaphysics, after the work of John Barrow and Frank Tipler. The metaphysics built on the science of any period in history is not neutral or value-free, but is an attempt, sometimes more honest than others, to locate human meaning in the results of science. Hence both the reductionist metaphysics of biologists like Richard Dawkins and Stephen Jay Gould, or of the philosopher Daniel Dennett, and the anthropic metaphysics of Fritjof Kapra, Gary Zakov, John Barrow and Frank Tipler (to name just a few on both sides), are highly valuable. Ever since Plato poured scorn on the Pythagoreans for attempting to understand musical intervals by listening, most Western philosophers have been remarkable for their lack of interest in the perceptual world, Kant another prime example. While the philosophers may have sharpened our ability to reason about abstract principles, the new metaphysics offers much more interest because of its roots in the measurable universe.

An individual who epitomises the early move to observation is Leonardo da Vinci, stating in his notebooks that he had little time for re-hashing the ideas of the Greeks. His thinking, though broadly medieval, was rooted in a powerfully developed gift for observation of the natural world, a gift that was as much artistic as scientific. Leonardo was also a consummate technologist. He worked a full century before the ideas of Galileo became known, but represented that part of the Renaissance that looked to the future (i.e. the Enlightenment), while the neo-Platonists, including Michaelangelo and Ficino's Academy, represented that part of the Enlightenment that looked back to the Greeks. Both were essential of course, and both operated under the medieval world-view summed up in the phrase 'as above so below', or in another common formulation: 'man as microcosm'.

Which brings us back to the anthropic metaphysics, introduced above, but not defined. Barrow and Tipler's formulation of the strong anthropic view is this: 'consciousness is as essential to the existence of the universe, as the universe is to the existence of consciousness'. The symmetry of this statement restores man in part to the place he had during the ancient world up to the time of the Enlightenment, in other words up to the time of the reductionist metaphysics of Laplace. However, the anthropic metaphysics was bound to follow on the heels of the first paradox facing science, that of the nature of light. Although the use of light (in telescope and microscope) had been central to the Newtonian physics, Newton could not do more than speculate on its nature. Later science showed that light was both a particle and a wave, depending on how the experimenter approached it. Even today most scientists resist the implication that the experimenter determines the outcome of the experiment, because this undermines the notion of the impartial observer. However, light has consistently forced this paradox on its investigators, firstly through the wave/particle duality, secondly because its speed is invariant with respect to the observer, and finally through the paradox demonstrated in the Schroedinger's cat gedanken experiment. Whatever a scientist may think of the anthropic metaphysics, as a social phenomenon it is on a significant scale. It also forces a re-evaluation of the medieval view under which Leonardo worked and wrote. Stephen Jay Gould has attacked Leonardo for pursuing the idea of 'as above so below' in Leonardo's parallel, extensively pursued, between the human circulation and the movement of water in streams, rivers and oceans. Gould has a point, but we don't ridicule Einstein for his hidden variables theory or for the debacle over negative gravity, so why attack Leonardo for a folly on a much smaller scale? The answer must lie with the reductionist rejection of the medieval principle, a principle now resurrected in the `anthropic' form.

This is all by way of a context for the main point being made here. The new metaphysics is a watershed in Western intellectual history, and while I side squarely with the anthropic side of it, I find the genetic reductionism of the Darwinists intellectually stimulating and vital to healthy debate. What I don't want to see is a metaphysics that becomes as speculative as that of earlier and generally redundant philosophical thinking, simply because it is not taking technology seriously. To understand this point we need to look at Tipler and Deutsch's ideas in more detail.

Frank Tipler in his Physics of Immortality postulates a time in the far future called the Omega Point when all of humanity will live in virtual reality, an artificial and enhanced simulacrum of the universe maintained by a computer of infinite calculating power. Presumably my imaginary students of the far future would be employed flat-out to build this virtual world, though Tipler implies that we could not just simulate but improve on reality. David Deutsch in his Fabric of Reality also requires a machine that can generate any world, though he is more cautious in predicting future computer power. Deutsch suggests that rather than computers speeding up in an exponential growth of calculating power, we slow down the metabolism of the brain. The acknowledged down side of this approach is that five minutes spent in the virtual world would be at the cost of five years suspended animation in the physical body! Tipler and Deutsch use virtual reality scenarios to construct different metaphysics however (though some would call it more like science fiction), Tipler for the proof of the existence of God, and Deutsch for a more interwoven cosmology involving parallel universes. In both cases, there is more scope for careful study of technology before leaping into the dark.

Going back to the technology of virtual reality, let us consider what would be involved in an attempt to create the virtual fragment used to introduce this essay. Firstly, an acutely developed artistic eye is needed, nothing short of the sensitivities of a Leonardo. Secondly computing power is needed in two aspects to deliver the virtual world, known as modelling and rendering. Modelling refers to the representation of the virtual world in the computer, and its evolution in time. Perhaps based on the relative cheapness of computer memory in recent years, both Tipler and Deutsch assume that memory will become effectively infinite, and so the modelling is of no difficulty. The rendering of this world is a different problem, and relates to the need to deliver sense impressions of it to the VR inhabitant. A visual perspective virtual cinematographic view has to be generated in real time, along with other sensory input, chiefly sound, aroma, and kinaesthetic data. As already mentioned Tipler assumes infinite computing resources available for this, though Deutsch accepts that there would be limits. The question then arises, are there limits to computing, and if so, do they fall short of delivering a believable virtual world?

For me, one virtue of teaching and researching 3D computer graphics is to force the virtual reality modeller to look again at the natural world. In doing so a profound respect for it is generated, a respect somewhat lacking in the theorists. The more one attempts to model natural phenomena (and there is a vast research effort going into this around the world) the more difficult the problem seems. The sheer richness of the visual world makes reconstructing the canal scene not just a highly ambitious engineering project, but perhaps simply impossible. An analogy that might be useful is the early optimism about colonising the moon. Are we applying the same optimism to the growth of computing power that we did to space technology in the sixties? Looking at virtual reality systems of today it is truly an open question. It has been suggested, and I have sympathy with this view, that to faithfully model a universe you would need a computer the size of a universe. There is no doubt that we can model (to use this term in a slightly different way) aspects of the real world for simulation, for artistic, leisure, and research purposes. I can think of three examples, all of which have cost a million dollars or more to implement, and which achieve their goals to a considerable extent: an installation by artist Char Davies called Osmose, the `hemispherium' at Teeside University used amongst other things to simulate town planning projects, and the planetarium in New York which will use computer-modelled starfields. In each case, the gap between the real and virtual versions is stark, to say the least, though this does not make the projects fail in their own terms.

To conclude, the recent phenomenon of the new metaphysics, that is philosophical and speculative writings with a deep root in science, is a major breakthrough in human thinking. It is at its strongest when its scientific basis is strongest, though even the wildest speculations in this genre may hold more value than some of the philosophies of old, divorced as they were from systematic, careful and sensitive observations of the natural world. This is also a plea however to take technology seriously, the same technology without which science could not arise. Computer technology is as intimately bound up with contemporary science, as the telescope and related instruments were with early science, but we should not make assumptions about its possible future development. Unless of course we pass a door in a wall, or turn our head too quickly, and find nothing.