Visions: How Science Will Revolutionize the 21st Century.

Visions begins on a note of arrogance. Unlike previous efforts to chart the future of technology, Michio Kaku assures us, his predictions are likely to be correct. As science approaches a full understanding of the laws of nature, a scientific consensus is emerging about where technology is headed and on what timetable. This book, Kaku asserts, reflects that consensus.

Baloney. What is remarkable about many of the advanced technologies Kaku discusses - artificial intelligence, genetic engineering, nuclear fusion, electric cars - is the distinct lack of scientific unanimity about their potential. For every physicist who says that fusion is "the energy of the future," there's another who replies, "Yes - and it always will be." Even when the experts are in general agreement - as they once were about the infeasibility of cloning an adult sheep - consensus has hardly proven a guarantee of predictive accuracy.

Nonetheless, Kaku, a theoretical physicist and high-profile popularizer of science, has written an absorbing book, filled with thoughtful speculations about the 21st century and beyond. Visions sketches what might emerge from three 20th-century scientific upheavals: the "computer revolution," the "biomolecular revolution," and the "quantum revolution." These revolutions are interconnected, as Kaku notes; discovery of the DNA double helix, for example, relied on X-ray crystallography, a technique derived from quantum physics. Such linkages, he expects, will take on growing importance in the next century, in the form of DNA-based computers and other hybrid technologies.

Visions provides an intriguing (and explicit) rejoinder to The End of Science, the 1996 book in which journalist John Horgan argued that the era of scientific discovery is sputtering out in disappointment and confusion. Similar to Horgan (but unlike eminent scientists such as Roger Penrose and Freeman Dyson), Kaku believes that breakthrough insights into nature's workings, such as evolution and relativity, are now mainly things of the past. But where Horgan detected intellectual drift and technological stagnation, Kaku sees something very different: The age of discovery is giving way to the age of mastery. Having learned the universe's rules, humans are finally ready to become full-fledged players in the game.

The quantum, biomolecular, and computer revolutions, in other words, are enabling us to be "choreographers of matter, life, and intelligence," no longer mere passive observers of nature's dance. Yet even while taking this expansive view of technology's potential, Kaku is adept at recognizing technological hurdles and limits.

Computing power, Kaku expects, will become increasingly cheap and ubiquitous in the next two decades, manifested in such products as wearable computers, smart cars, and digital scrap paper. Helpful (but sometimes annoying) "intelligent agents" will sort your e-mail, update your schedule, and remind you to watch your diet. But before long, Kaku notes, chip making will bump up against the physical limits of silicon, and further progress will depend on the development of holographic memory, organic processors, quantum transistors, and other exotic technologies.

After 2020, Kaku predicts, the first glimmerings of true artificial intelligence will appear, as computers acquire common sense and as the Internet evolves into something similar to the "magic mirror" that imparts wisdom in fairy tales. After 2050, robots endowed with some degree of consciousness and self-awareness may roam the earth. Might humanity eventually be enslaved or slaughtered by its robotic creations? Kaku closes his discussion of artificial intelligence with an overview of the built-in safeguards that should be devised to prevent such an outcome.

Biotechnology also will make vast strides in the early 21st century, according to Kaku. By 2020, the genetic underpinnings of many hereditary diseases will be understood, and entire classes of cancer will be curable. People will own CD-ROMs containing their own personal DNA codes. Between 2020 and 2050, genetic research will see slower progress, as scientists grapple with the intricacies of gene function and protein folding. During this period, however, it will become possible to grow new vital organs in the lab, perhaps extending the human life span by decades.

After the century's midpoint, Kaku writes, "we may be able to manipulate life itself." Yet he is impenetrably vague about what this means. More interesting is Kaku's discussion of feats that probably lie beyond biotech's reach. Performing major design changes on human beings - say, growing wings on a person's back, in Kaku's whimsical example - is unlikely to be feasible even in the late 21st century. Consider the obstacles involved: The genes that initiate wing formation in a bird or insect may do nothing in a human (or may activate homologous organs, such as arms); these genes would have to be altered to allow a wingspan of some 20 feet; and the human's entire genome would have to be transformed to create the lighter bones and stronger muscles required for flight.

Surveying the "quantum future," Kaku assesses a broad range of possibilities for manipulating matter and developing new sources of energy. Electric cars and magnetic-levitation trains are emerging as viable forms of transportation, he argues, and solar power is poised to become a leading energy source. Room-temperature superconductors and microscopic lasers may find numerous industrial applications. Nanotechnology's molecule-sized machines are of uncertain feasibility, Kaku notes, but dust-sized sensors and motors will be used widely in the coming decades. Some other staples of science fiction, such as force fields and portable ray guns, appear to be incompatible with known laws of physics, he adds.

Space technology will make steady, if unspectacular, progress in the next few decades, according to Visions. Kaku is dismissive of the notion of a manned mission to Mars in the early 21st century, basing his argument on exorbitant cost estimates now widely regarded as erroneous. Yet after 2020, he emphasizes, astronomical instruments may be sensitive enough to detect Earth-like planets in other solar systems. The century's latter half may see ambitious efforts to develop fusion-powered interstellar rocket ships. Antimatter engines loom as an intriguing prospect sometime beyond 2100.

Kaku's social and political asides are less imaginative than his technological speculations. He argues, plausibly but predictably, that the economic strength of nations in the 21st century will depend on their technological prowess. In chapters devoted to "second thoughts," he presents grim scenarios of "information ghettos" and bioengineered germ weapons; his solutions are unremarkable generalities about education and international cooperation. Some of Kaku's political pronouncements are mere cliches. Discussing the nation-state's future, he writes, "As John Lennon said in his song 'Imagine,' perhaps it's not hard to imagine a world without nations."

Might a public backlash against technology derail much of the progress forecast in this book? The history of advanced technologies in the 20th century - nuclear energy comes to mind - indicates that not everything that is technically feasible will end up receiving political and social acceptance. Certainly, the 21st century's "choreographers of matter, life, and intelligence" will face their share of protest movements and hostile regulators. Visions, however, has little to say about such matters.

That is unfortunate, since Kaku's own experience might have provided an interesting perspective. A longtime antinuclear activist, he was a leading figure in the 1997 protest campaign against the Cassini space probe, a plutonium-using scientific mission to Saturn. Critics of the anti-Cassini movement, including me, argued that the campaign relied on gross exaggerations of the mission's risks and that the broad opposition to "nukes in space" threatened to cripple space exploration. In addressing the uncertainties of technological change, Kaku the author might have taken some tips from Kaku the activist.

Yet even if technology follows a more unpredictable - and politically volatile - path than the one glimpsed in Visions, the book's strengths readily outweigh its weaknesses. Kaku's predictions are intelligent and thought-provoking, and his technological optimism never veers into an unconvincing techno-utopianism. Moreover, no one can accuse him of thinking small. Looking beyond the 21st century, Kaku sketches out a bold future of galactic colonization and more.

Drawing upon categories devised by Russian astronomer Nikolai Kardashev, Kaku sees technological civilizations advancing through several phases: Types I, II, and III. Type I refers to a global civilization, the masters of a single planet. A Type II civilization utilizes the resources of an entire solar system; such a society might even build a vast shell or "Dyson sphere" around its star. A Type III civilization operates on a galactic scale, occupying numerous solar systems. In this scheme of things, humanity is currently a backward society, or Type 0, but is on the verge of attaining Type I status.

Becoming a Type II civilization will take many centuries, according to Kaku, and achieving galactic Type III status requires many millennia. But it's worth the effort: Civilizations of Types II and III are invulnerable to asteroid impacts, supernova explosions, and other natural disasters.

Toward the book's end, Kaku launches into a discussion of wormholes, super-strings, and other exotica of modern cosmology. After billions of years, even galactic civilizations are doomed, as the universe freezes in a Big Chill or collapses in a Big Crunch. But on the book's last page a new category is introduced: the Type IV civilization, masters of space and time. Such beings might be able to-build tunnels to parallel universes. Here, then, is Kaku's ultimate statement of technological optimism: Intelligent life might survive the end of our universe.

Kenneth Silber ([e-mail]) writes about science, technology, and economics.