"This is called a magic mirror in Japanese, as it is in Chinese. It was known in China in the ninth century and is mentioned by Needham in his definitive work on Chinese science and technology," said the number-two antique dealer in Kyoto to me in the early 1970s. The one I bought, for $600, of the two he had, was about about 9 inches in diameter and made of cast bronze, with Chinese ornamentation on one side and a very tarnished mirror surface on the other. In the dark, ancient, tatami-floored shop, this early Edo mirror was held up to the light of northern Kyoto sky throught shoji windows, and the reflection it projected on the 400-year-old soot-darkened pine wall boards, was of a standing Amida Buddha statue with a luminescent halo. (The mirror I didn't buy had a sharper image of three vertical Chinese characters used in sacred chanting.)

I brought the mirror home to San Francisco and put it in my living room. I found that there were no visible impressions in the green tarnished bronze side and no detectable surface or sub-surface impressions on the mirror side. The mirror, which could be illuminated with either electric light or sunlight, would cast an image, at any distance, of about the same size. This kind of mirror was customarily used in temples, where the image could be cast over a distance of fifty or more feet; my living room was less than twenty feet wide. The sun lit image is awesome. The clarity of the image remained the same at any distance, but the brightness diminished with the distance the light traveled. The image remained the same regardless of the distance to the source of light as well.

I have read about science all my life. I studied physics in a class taught by Niels Bohr, I am a devoted pilot, and I never stop experimenting and tinkering. Yet in my opinion, the phenomenon of this mirror could not be explained by scientific principles. If the Amida Buddha image were somehow embedded in the reflective surface, then the mirror must have a focal point. The image would then grow sharper or duller as the distance from the mirror changed or as the distance to the source of light changed. Try pasting large postage stamp stuck on a flat mirror, or on a concave shaving mirror. The sharpness of the stamp shape will change as the distance from the mirror changes and as the distance to the light source changes. With the Japanese magic mirror, this wasn't true. What was going on?

In my opinion, there was a holographic image on the surface of the mirror, since a holographic image will not change sharpness with distance. However, holograms can only be made with coherent light, which today is only available from lasers. And a holographic image as seen on a film negative is a wavelike pattern like the surface of water&emdash;visually meaningless. The explanation is sound, but the mechanism is not known.

Two years later, I went back to Japan and, with the aid of good academic friends who have great patience, I located a technical article on the mirror that confirmed the characteristics of the mirror as I have described them. The article offered no explanation, but showed that under the mirror's reflective surface, when viewed through an electron microscope, there was great variation in the surface of the cast bronze at the level of angstrom differences. Although there was no recognizable pattern, the variations would create refracted light.

My hypothesis is that over a period of hundreds of years, a holographic image could have been developed through pure experimentation, and some rules for such imaging could have been conceived by the early Chinese. Imagine painting dots and lines on your test mirror (the one you put the postage stamp on) for a whole lifetime or several lifetimes to try to achieve an image that didn't loose its sharpness with distance. That is how I think the magic mirror technology was developed.

I checked out one more detail while in Kyoto. I found a craftsman who maintained magic mirrors. Working in his home, he gave my mirror a new surface in two hours. He first stripped off the mercury down to a nearly bare surface, then reapplied it with old rags, pear seed oil, and a continually rotating surface motion similar to that used to polish astronomical lenses. He did this while sitting cross-legged on the floor, with the mirror on a pad on his knee, surrounded by open jars of oils, buffing compounds, and polishing compounds.

The newly surfaced mirror, lacking the old tarnish and the fingerprints around the edge, was a little brighter in reflecting the image, but there was no increase in the sharpness of the image at any distance, as my theory would have predicted.

I wrote to many science and technology institutions, sending the Japanese technical article and its translation with an offer to donate the mirror. There were no responses, so I gave it to Frank Oppenheimer of the San Francisco Exploratorium, where it has since disappeared.

This mirror makes an important point about science and technology. Technology is not just ahead of science, it's a driving force, as is apparent in such fields as super-conducting materials, where new materials and super-cool phenomena occur before theory can explain them. In molecular biology, new measuring instruments (x-ray crystalography and nuclear magnetic resonance) create new theories. This is also true in other branches of science, though it is harder to see.

Moreover, because this mirror was a millenium ahead of science, it suggests that technology has opened a vast field of data and observations that science is slowly exploring only a few square yards at a time.

Science lets us see, and at the same time it keeps us from seeing. Technology is just a blind set of ad hoc rules that diddles with the empiric world. Science offers explanations that allow us to understand what we have seen, as in the case of Galileo, who saw the moons of Jupiter and so realized that the earth was not the center of the universe.

On the other hand, Western astronomers didn't record supernovae for centuries, because they didn't believe in such transitory phenomena, while the Chinese did record them because they were within the Chinese view of existence.

A metaphor that seems to shed light on the subject is the jigsaw puzzle. There are two ways that puzzle pieces can be combined, as we all have learned by age 12. One is to look for connecting shapes on the table and try to fit them together. When there are clusters of connecting pieces on the table, some will evoke a design or color pattern that allows us use the second method: adding many extra pieces by selecting deductively. Step one is analagous to technology, step two to science.

I abstract three propositions from this metaphor: One is that some cultures with long histories of technology will do well in the current world, where they can rely more on technology and less on science. The obvious example is Japan, but other examples are China and West Africa. China developed technology to great heights very early in its history, which makes it hard to understand why she governed herself so poorly in later centuries.

West Africa also has a history of technological development. When Europeans began exploring Africa, they found ironwork in this area of such sophistication that we now recognizeit steel. In fact, the only market for European iron in Africa was for those who wanted to melt it down for their local hearths to make into steel. The same was true of cloth fabrication at the time: African weaving was far ahead of anything the Europeans had, and the Africans disdained their imports. In the same century, the Japanese had steel of such high quality that it could be used to chop European metal.

My second proposition is that we should not be surprised if new areas of technology appears in unlikely places. The magic mirror is but one example; others could be found in medicine (the ancient pharmacopia was especially effective), agriculture (such as that practiced by so-called hunter-gatherers), and animal husbandry (such as beekeeping and silk production). Speculatively, I would add to these weather control (the Tibetans claim to be able to do it), mental and physical health, and social behavior.

Lastly, I propose that science can dissolve into theoretical mush on the boundaries of its efforts. This is most apparent in physics, but it will become more evident in other fields as they approach the sophistication of quantum mechanics. That is, when you know a pattern for a small area of a puzzle, it will carry you just so far in selecting other pieces based on color or design before its usefulness expires.