What Is Life? Summary and key ideas

What Is Life?, by Erwin Schrödinger.

Schrödinger opens with a problem that belongs to physics, chemistry, and biology at once. Modern specialization is a paradox: reliable facts are accumulating, yet each specialist can command only a small part of the whole. He presents his synthesis cautiously, aware that crossing disciplinary boundaries means relying on second-hand knowledge and incomplete expertise. He also keeps the mathematics selective, since the argument is meant to connect fields rather than reproduce every technical derivation. The guiding question is how physical and chemical events inside a living organism can be explained by physics and chemistry, especially when orderly biological effects may depend on scales smaller than ordinary statistical reasoning expects.

That difference becomes clearer when we think about scale. Asking why atoms are so small is, in part, asking why organisms must be so large compared with them. Our practical units of length reflect the scale of the human body, so atomic dimensions feel impossibly tiny. Yet the more important issue is not a choice of units. It is that our bodies, senses, and instruments are built from vast numbers of atoms, while the events that matter in life may be controlled by much smaller groups.

Human senses cannot register individual atoms directly. The organs of sensation are too coarse, and too many atoms participate in every perceptual event. If a single atomic impact could alter experience in an uncontrolled way, perception would become chaotic. In such a world, orderly thought would be impossible, including the thought that atoms exist. Because thought is orderly and applies to orderly perceptions, Schrödinger argues that the brain and sensory system must obey physical laws to a high degree of accuracy. The interactions between an organism and its surroundings must also be orderly enough to support that stable experience.

Physics usually obtains this reliability statistically. Individual atoms are in continual heat motion, so a small collection is disturbed by accidents that have no stable pattern. When enormous numbers cooperate, irregularities partly cancel, and an average becomes dependable. The laws that describe relevant physical and chemical processes are therefore statistical in character. Other regularities might exist in principle, but thermal motion would continually disrupt them unless enough particles shared in the process.

Paramagnetism offers a compact illustration. Oxygen molecules behave as tiny magnets. An applied field tends to align them, while heat tends to randomize their orientations. Individual molecules do not all point in the same direction, but the average magnetization is predictable and, within the ordinary range described, proportional to the field. Lowering the temperature weakens the opposing thermal motion, so magnetization increases. At sufficiently low temperature, it approaches saturation instead of continuing to grow proportionally. These are presented as experimentally confirmed features. The crucial point is that the observable magnetization belongs to a large population. With only a few molecules, the result would fluctuate rather than settle into a steady law.

Brownian movement makes the same principle visible. A fog droplet may sink at a regular average rate, yet its path is continually knocked about by impacts from surrounding molecules. Very small organisms can be affected by the same thermal agitation; unless their own locomotion overcomes it, their movement is partly at the mercy of the medium. Randomness does not prevent a large-scale average, but it becomes harder to ignore as the number of participating particles falls.

Diffusion gives another example. A dissolved molecule does not need to be pushed away from a crowded region by a special repulsive force. Each molecule moves independently, receiving random impacts. Still, a net movement toward lower concentration appears because more molecules cross a boundary from the side that contains more of them. The diffusion equation captures this average pattern in an exact mathematical form, but its physical application remains approximate because it describes chance events. The approximation improves when enormous numbers cooperate and becomes visibly erratic when fewer molecules are involved.

The same thermal noise limits measurement. In a torsional balance, molecular impacts can disturb the light body suspended from the instrument. The disturbance sets a practical limit on how finely the device can register a signal. Repeated observations are needed so that Brownian movement can be averaged out. Sensory organs face the same constraint. Excessive sensitivity would not necessarily improve perception; thermal noise could overwhelm the signal and make the organ useless as an instrument.

The size of the statistical error follows a simple rule. If n molecules cooperate, the probable relative error is of the order of one divided by the square root of n. With one hundred particles, a departure of roughly ten percent is still expected. With one million, the corresponding error is roughly one part in a thousand. More particles do not remove molecular irregularity. They make its effects smaller relative to the average.

This logic suggests an older physical expectation: an organism and its important processes should have a comparatively gross structure so that enough particles cooperate to produce accurate laws both inside the body and in its exchanges with the environment. Schrödinger says that expectation is only partly right, and that biological knowledge reverses its most important assumption. Very small atomic groups can dominate orderly events in an organism and determine large-scale developmental features or important functional characteristics.

That is the central challenge. Physics normally explains dependable behavior by letting vast populations average away individual accidents. Life can make a small, specifically organized structure physically consequential. The difficulty is not evidence that living processes lie beyond physics. It is a demand for a broader account of physical organization, one that explains how order can remain effective when particle number alone should offer too little statistical protection. Having established this scale-and-statistics baseline, Schrödinger is ready to ask what kind of material structure could carry such organized effects.