第4章
Shadows of the Multiverse
I'll demonstrate quantum physics through simple shadow experiments rather than using complex equipment. When an electric torch shines in a dark room, its light spreads in a cone. But light isn't infinitely divisible-it comes in discrete packets called photons. At great distances, observers don't see dimmer light but rather individual flickers of constant brightness as photons arrive less frequently.
When light passes through small holes, it creates complex patterns rather than sharp shadows. With two narrow parallel slits, light creates an interference pattern of alternating bright and dark bands-nothing like the silhouette of the barrier casting the shadow. These strange shadow patterns reveal fundamental properties of light that lead to extraordinary conclusions about reality.
Even more astonishingly, this interference occurs when photons pass through the apparatus one at a time. Each individual photon behaves as though it's being influenced by something coming through the other slits, even when there are no other photons present. This leads to an inescapable conclusion: each photon must be accompanied by invisible "shadow photons" that pass through the other slits and interfere with it.
These shadow photons behave exactly like regular photons-they travel at light speed, bounce off mirrors, refract through lenses-yet they cannot be directly detected. They form a vast, invisible reality alongside our tangible world. This phenomenon isn't unique to photons but occurs with all particles-electrons, neutrons, everything. Our tangible universe is merely one among countless parallel universes, each containing shadow particles that interact with each other just as ours do, but which affect our universe only through interference.
Some physicists ask why we can't simply say photons behave "as if" interacting with invisible entities without claiming these entities are real. But this view inevitably leads to nonsense. A tangible photon behaves differently depending on what paths are available elsewhere for something to travel along. Something must travel those paths-the possible cannot interact with the real; non-existent entities cannot deflect real ones. If a photon is deflected, something must have deflected it.
There's no privileged "tangible" universe among the many parallel universes. What I call "tangible" photons, my counterparts in other universes call "shadow" photons, and vice versa. I may subjectively feel distinguished as the "tangible" one because I directly perceive myself and not others, but all my counterparts feel the same about themselves. Many versions of me are writing these very words right now-some putting it better, others having gone for tea.
第5章
Knowledge Through Problem-Solving
What justifies inferring grand conclusions from scant evidence? We don't directly perceive external reality but only electrical impulses in our nerves-patterns of weak current trickling through our brains. Yet we interpret these as evidence of a vast, complex multiverse.
This evidence can't logically prove external reality exists at all. Solipsism-the theory that only one mind exists and external reality is merely that mind's dream-cannot be logically disproved. Since any observational evidence is consistent with infinite possible theories, we can't logically deduce anything about reality from observation.
Inductivism claims scientific knowledge works by generalizing observations: we observe phenomena, induce theories from patterns, verify these theories through repeated observation, and eventually become convinced of their truth. But this conception is profoundly false. Theories aren't mere generalizations-did we observe one universe, then another, before inducing there are trillions? Repeating observations rarely convinces us of theories. And as Russell's chicken story shows (the chicken that predicted the farmer would always feed it until the day he wrung its neck), induction cannot justify conclusions.
Karl Popper's theory views science as a problem-solving process. Unlike inductivism, scientific discovery needn't begin with observation but always begins with a problem-ideas that seem inadequate and worth improving. We solve problems by proposing new theories (conjecture), then subjecting them to criticism, comparing which offers better explanations. When a theory fails criticism, it's abandoned. If we replace original theories with better ones, we've made progress-though tentatively, as even these will likely be altered or replaced.
Science seeks not eternal truth but the best explanation available now. This process resembles biological evolution-both create knowledge through variation and selection, with science doing so more efficiently through conscious criticism rather than natural selection's blind process of death and reproduction.
第6章
Criteria for Reality's Existence
Having rejected solipsism, we need criteria to determine what is real in competing explanations. When something behaves as if it existed by affecting our observations, we have evidence it exists. What matters isn't direct perception but a phenomenon's role in our best explanations.
Even powerful "kicks" may not convince us of something's reality if better explanations exist-like waking from a nightmare about monsters. Conversely, we might rationally infer the existence of an unseen prankster with an air-gun based on indirect evidence if it's the best explanation available.
The complexity of an entity is particularly relevant-we prefer explanations that account for complexity without introducing unnecessary complications. Dr. Johnson's criterion can be rephrased: "If, according to the simplest explanation, an entity is complex and autonomous, then that entity is real." This criterion explains why we accept planets as real but not mirror reflections, which merely borrow their complexity from elsewhere.
Computational complexity theory helps us understand what makes entities real. If substantial computation would be required to create the illusion of something, that entity is likely real. This explains why shadow photons in quantum interference experiments must exist-calculating how photons react to distant objects requires the same computational effort as tracking shadow photons through their interactions.
Reality is saturated with evidence that's accessible to anyone who looks properly, regardless of location or time-a science-friendly property Galileo recognized. Physical reality shows self-similarity across scales, with patterns repeating throughout the universe and multiverse, creating the self-similarity we call knowledge.
第7章
Virtual Reality and the Nature of Experience
Flight simulators exemplify virtual reality technology, allowing pilots to experience flying without leaving the ground. They can be programmed with characteristics of real or imaginary aircraft and environments, providing visual images, physical sensations, and instrument readings that respond appropriately to the pilot's actions.
While these are rudimentary virtual-reality generators, true virtual reality requires both wide sensory coverage and substantial interaction between user and simulated entities. Modern systems typically use helmet-mounted displays with headphones and special gloves with sensors and effectors, allowing users to see, hear, and feel a responsive simulated environment that changes appropriately as they move and interact with virtual objects.
Virtual reality doesn't invalidate Dr. Johnson's refutation of solipsism. When a flight simulator "kicks back" in response to a pilot's action, what's responding isn't an engine but a computer running calculations-yet these calculations are real physical processes occurring in a real physical object. Rather than demonstrating human limitations in understanding reality, virtual reality actually reveals our unlimited capacity to comprehend the world.
Once we understand the neural codes our sensory organs use to communicate with the brain, we could directly stimulate the relevant nerves, eliminating the need for physical simulations of gravity, chemical scents, or other difficult-to-reproduce phenomena. This approach would enable "feelies"-Huxley's term for multi-sensory experiences-allowing us to feel ocean waves, smell sea air, and experience sunset winds without leaving home.
The true heart of virtual reality is the computer that simulates the environment. The sensors and image generators are merely the interface-the "connecting cable" between user and computer. Once we solve the finite problems of intercepting nerve signals and cracking neural codes, virtual reality's focus will shift entirely to programming environments rather than building better interfaces.
Every virtual-reality generator, even when simulating impossible environments like faster-than-light travel, is simultaneously rendering some physically possible environment-itself. This creates a profound connection between virtual worlds and physical reality: any environment renderable in virtual reality must be perceptibly indistinguishable from some physically possible environment.
Most profoundly, all our "direct" experience of the world is actually virtual reality-our unconscious minds generate renderings from sensory data and interpretive theories. Every scrap of our knowledge, from physics to philosophy to fantasy, is encoded as programs for our brain's virtual-reality generator. This relationship with virtual reality is what defines humanity's ecological niche-virtual-reality rendering is the characteristic means by which humans survive.
第8章
The Universal Computer and Reality's Self-Similarity
At the heart of virtual reality lies computation, which raises the question of what environments can be rendered. While computers with unlimited memory capacity can be envisaged through mechanisms that read unlimited numbers of disks, computers with unlimited speed cannot-they will always have fixed maximum speeds. This creates a fundamental challenge: how can a virtual-reality generator accurately render highly complex environments when calculating a single second of experience might take longer than a second to compute?
The ultimate virtual-reality generator would need to control the processing speed of the user's brain. Since nerves can only fire at certain rates, the computer has a small window to decide when each nerve should fire next. When calculations require more time, the computer would slow down or stop the brain until computations are complete, then restore normal speed.
Despite our hypothetical advances in virtual reality technology, there are fundamental limits to what any physically possible virtual-reality generator can render. Using a diagonal argument similar to those employed by Cantor, Godel, and Turing, we can prove that for every environment in a virtual-reality generator's repertoire, there exist infinitely many "Cantgotu environments" that cannot be rendered.
While we cannot render all logically possible environments, we can create a universal virtual-reality generator-one whose repertoire contains that of every other physically possible virtual-reality generator. This requires a universal computer capable of calculating anything that can be calculated. The Turing principle states that there exists an abstract universal computer whose repertoire includes any computation any physically possible object can perform.
This remarkable property means a single, buildable physical object can mimic all behaviors of any other physically possible object or process-making reality comprehensible and enabling the evolution of living organisms. The laws of physics thus mandate their own comprehensibility, as they make it physically possible for those same laws to become known to physical objects.
第9章
Life's Significance in the Multiverse
Life is a fundamental phenomenon of nature, though not for the reasons Aristotle thought. While ancient philosophers believed animate matter had special properties like self-initiated motion, modern biology reveals life isn't about special physical substances but about replication. Life on Earth is based on molecular replicators called genes-sequences of DNA that function as computer programs. These genes cause certain environments to copy them through layers of complex control and feedback mechanisms.
A replicator must causally contribute to its own copying-its specific physical form makes a difference to whether copying occurs. Genes accomplish this by programming cells to manufacture specific proteins that enable the organism to survive and reproduce. Unlike "junk DNA" which gets copied regardless of its sequence, changing a functional gene like insulin would prevent copying by causing the organism's death.
Living processes are remarkably similar to virtual-reality generation. Cellular "computers" execute gene programs, exerting precise control over a complex environment (the organism) to make it act back upon the genes in a specific way-replicating them. The organism is both rendered and manufactured according to specifications in the genes.
Despite appearances, life is not insignificant in its physical effects nor theoretically derivative. Though life appears physically negligible, its true significance lies in knowledge. Life is about the physical embodiment of knowledge-entities are adapted to their niches if they embody knowledge that causes the niche to keep that knowledge in existence.
Throughout the galaxy and multiverse, stellar evolution depends on whether and where intelligent life has evolved. Cosmologists have found that surviving life would eventually make major changes to galactic structure and later to the entire universe. We've underestimated life's physical impact by being too parochial, seeing only the past where life has affected nothing astrophysically significant. But the future may be dominated by life's effects.
There is indeed a fundamental physical difference between knowledge-bearing and non-knowledge-bearing objects, visible through the multiverse view. When cosmic rays cause mutations, they strike DNA differently across universes, creating different variations. The knowledge-bearing gene segment maintains the same sequence across most nearby universes because mutations that alter it would prevent replication. Meanwhile, the junk-DNA segment varies greatly between universes, potentially showing every possible sequence variation.
From the multiverse perspective, these segments aren't remotely alike-knowledge-bearing matter forms regular structures across universes, like crystals in the multiverse, while non-knowledge matter appears random and irregular. Life itself creates the largest distinctive structures in the multiverse, with living organisms maintaining recognizable forms across many universes while non-living structures like constellations vary completely.
第10章
Quantum Computers: Collaboration Between Universes
Quantum computation isn't merely a faster technology for implementing Turing machines-it represents a wholly new way of harnessing nature. A quantum computer uses uniquely quantum-mechanical effects, especially interference, to perform computations impossible even in principle on any classical computer. It's the next evolutionary step in technology: after tools powered by muscles, domesticated organisms, machines harnessing materials and forces, and classical computers processing information, quantum computers will be the first technology allowing useful tasks to be performed in collaboration between parallel universes.
Classical and quantum unpredictability differ fundamentally. Classical chaos stems from extreme sensitivity to initial conditions-tiny measurement errors grow exponentially, making prediction impossible. But real systems follow quantum mechanics, where small deviations cause only small prediction errors. Instead, quantum unpredictability comes from the multiverse nature of reality-systems spread across universes, appearing random from our single-universe perspective.
While quantum theory often makes probabilistic predictions, some quantum experiments yield definite, non-random outcomes through interference effects. An interferometer demonstrates this by using mirrors and semi-silvered mirrors to manipulate photon paths. A single photon entering the interferometer causes universes to differentiate when it encounters the first semi-silvered mirror-in half the universes it passes through, in the others it reflects. These versions later interfere at a second semi-silvered mirror, causing all universes to become identical again with the photon taking a specific path.
The complexity of quantum interference calculations increases exponentially with the number of interacting particles. With two particles each having a thousand possible paths, we must calculate a million different states. This exponential growth makes computing quantum system behavior truly intractable using classical methods, despite being perfectly predictable in theory.
In 1985, Deutsch proved that a universal quantum computer was possible by adapting Turing's constructions using quantum theory instead of classical mechanics. Such a computer could perform any computation that any other quantum computer could perform, rendering any finite physically possible environment without exponential resource requirements.
Quantum computers can efficiently perform mathematical tasks that are classically intractable. The most spectacular is Shor's algorithm for factorizing large numbers, which requires only a few thousand arithmetic operations across 10^500 interfering universes. When factorizing a 250-digit number, the computation occurs simultaneously across 10^500 universes that are initially identical but become differentiated within the factorization engine. This phenomenon demolishes the single-universe worldview-how else could such vast computations be performed when our visible universe contains only about 10^80 atoms?
第11章
The Nature of Time in the Quantum Multiverse
Time, despite its familiarity, has a reputation for profound mystery. Its basic attribute-that the present moment moves continuously toward the future-is pure common sense yet fundamentally nonsensical. The common understanding of time imagines it as a line with moments as fixed points, with a moving "present moment" sweeping from past to future.
When we try to illustrate the "moving present" consistently, we discover it can't actually move-at each moment, "now" is simply that moment. Nothing can move from one moment to another; to exist at a particular moment means to exist there forever. Our consciousness exists at all our waking moments, not as a single entity moving through time. Different snapshots of ourselves perceive different moments as "now," but they don't take turns being present or conscious-they are all conscious and subjectively in their present.
Common sense clings to two incompatible concepts of time: the moving present and the sequence of unchanging moments. When describing events, we think of time as unchanging snapshots; when explaining causality, we invoke a flowing present. This deep equivocation can't decide whether the present is objectively a single moment or many.
Physics achieved coherence by abandoning the flow of time. Despite Newton's poetic language about time flowing "equably," no physical theory has ever incorporated or been compatible with flowing time. Instead, physics treats time as a fourth dimension alongside space, creating what we call spacetime-the "block universe" where all of physical reality (past, present, and future) exists unchangeably in a single four-dimensional structure.
In spacetime physics, the future isn't open but fixed and unchangeable, just like the past and present. However, this contradicts our intuitive understanding of causation. For X to cause Y, two conditions must hold: both must happen, and Y wouldn't have happened if X had been different. But this reasoning involves counterfactual conditionals that have no meaning in spacetime physics, where only one thing actually happens.
The paradox dissolves when we understand physical reality not as a spacetime but as the multiverse-a vast collection of slightly interacting spacetimes. In the multiverse, counterfactuals refer to variants that actually exist somewhere, making it meaningful to say certain events are common or rare across universes. An event X causes Y in our universe if both occur here, but in most variant universes where X doesn't happen, Y doesn't either.
In the multiverse, any snapshot that exists is present in an infinity of identical copies. If I ask "which copy am I?", the answer must be "all of them"-assuming otherwise would require some external frame of reference that doesn't exist. Quantum theory doesn't determine what will happen in a particular snapshot, but rather what proportion of snapshots in the multiverse will have a given property. This explains why we can only make probabilistic predictions of our experiences (like coin tosses), even though what happens in the multiverse as a whole is completely determined.
Time is not a sequence of moments, nor does it flow. Yet our intuitions about causality, the open future and fixed past remain valid when properly understood in quantum terms. Our traditional theories of time fail because they try to express these intuitions within false classical physics. In quantum reality, we exist in multiple versions across universes called "moments," each version unaware of the others but with evidence of their existence through the physical laws that link different universes.
