Constructing a Theory of Life

What made you decide to click on this article? Perhaps you were googling, "how does intelligence arise from physical laws?" or "how did consciousness evolve?" or even "what is ’constructor theory’?"—referring to a recently developed framework that aims to encompass all of physics, and to explain the rise of sophisticated observers who can use Google.

Or maybe you were just browsing the FQXi website.

But according to fundamental physics, you didn’t actually make a choice. Your fate was encoded in the initial conditions of the universe—mysteriously fine-tuned so that life, you, and this article could arise. You’re shackled by the same physical laws as the universe in which you evolved; so there’s no such thing as truly ’free’ will.

Some scientists just don’t buy this standard story, however. At a time when we’re debating whether super intelligent machines will one day take over the planet, the question of what it means to be human is increasingly pressing. Many argue that free will is a precondition to being an autonomous and morally responsible agent. If free will could be shown to be an integral part of how nature works, rather than a social invention, that would definitely be a boost for human dignity.

Taking a step to rescue free will is Sara Imari Walker, a theoretical physicist and astrobiologist at Arizona Sate University, in Tempe. She says the notion that intelligent beings only have the illusion of influencing events in the unfolding universe through their choices, "agitates" her. With a grant of over $70,000 from FQXi, she is trying to come up with a theoretical framework, based on constructor theory, to explain how sophisticated observers can use knowledge to construct their own future, while still obeying the laws of physics.

Base Metal into Gold

There are many cosmic processes that can be created by intelligent beings, notes Walker. Just as a planet can form by the gradual accumulation of material, it can disintegrate bit by bit. This can happen ’naturally,’ but it can also happen as an intelligent civilization uses up the resources on its planet by building satellites or space stations, and placing them in orbit. Similarly, stars can transmute base metals into gold, but so can humans. In both cases, information in the form of knowledge is required for humans to carry out these processes. Yet conventionally, physics doesn’t recognize knowledge or information as a fundamental aspect of reality, let alone explain how observers can use it to become active participants in the world.

Yet humans are now able to use their knowledge to make elements never seen on Earth, adding new ingredients to chemists’ cookbooks. The periodic table was once filled only with natural elements that humans had discovered, which were forged in ancient cosmic furnaces, like neutron star mergers or supernova explosions. But there are also now higher-order elements, notes Walker, referring to substances such as Americium which can be created artificially and don’t occur naturally. "I think just having them on the same periodic table really says something about what’s possible in our universe if you have technological civilizations," she says.

Making artificial elements is one way that people tamper with the natural physical order. However, the fundamental role of observers, intelligent or otherwise, on constructing reality has long been a sticking point in physics. The rise of quantum mechanics, which governs the microscopic world of atoms and particles, has made the effect of humans on the physical world increasingly difficult to ignore. Under its strange laws, the act of observing a specific quantum system mysteriously affects how it behaves. This is not just a mathematical quirk, but something that’s constantly seen in experiments. But why should the presence of a human consciousness looking at a quantum system have such a profound effect on its behavior? That’s one question that Walker is trying to address.

Information is Physical

Walker believes it is time to take observers seriously—and that understanding information lies at the heart of the matter. (Her thoughts are outlined in her award-winning FQXi essay, "The Descent of Math.") To develop her ideas, she needed a ’meta theory’ that can encompass all physical theories, and in which the roles of the observer and information can be illuminated.

Enter constructor theory: Proposed in 2012 by David Deutsch, a quantum physicist at the University of Oxford, UK, constructor theory strives to fully explain the physical world by stating only which transformations are possible, which are not—and why. Entities that have the ability to carry out such transformations accurately and repeatedly are called "constructors." A kettle connected to power, for instance, is a constructor that can carry out the task of heating water given a sufficient amount of energy (D. Deutsch, arXiv:1210.7439 (2012)).

A crucial aspect of constructor theory is that it takes information to be fundamentally physical—determined by physical laws—rather than an abstract, mathematical concept. Chiara Marletto, a quantum physicist at Oxford who developed the theory with Deutsch, and is Walker’s co-investigator, says it is already proving useful. For example, she and Deutsch recently used the theory to describe information processing in a way that unifies the classical world of macroscopic objects and the quantum realm, without needing to ascribe magical status to human consciousness to explain why it disrupts quantum systems through observation (D. Deutsch & C. Marletto, arXiv:1405.5563 (2014)). "It’s basically a new way to formulate the laws of physics," explains Marletto.

The pair of physicists achieved this unification by stating that a characteristic trait of an object containing information is that it has certain physical properties that can be copied. (For example, the colour of a flag can be copied, and this feature is used to signal in air-traffic control.) In the classical word, all properties of physical systems can, in principle, be copied to arbitrarily high accuracy, so that they can be passed from a transmitter to a receiver. By contrast, quantum physicists already know that in the quantum world, it is impossible to copy or clone a quantum state. So there are properties of quantum systems that cannot be copied to arbitrarily high accuracy by the same machine.

Marletto and Deutsch showed that, using solely the principles of constructor theory, including this constraint on copying in quantum systems, it is possible to derive all of the qualitative properties of quantum systems, including the notorious fact that an observer measuring a generic quantum state will inevitably disturb it. This shows explicitly that a simple constraint on copying information can help solve the puzzle of how observers (which are usually taken to be nothing more than simple detectors, in the framework) disrupt quantum systems by monitoring them.

Now, Walker and Marletto want to move up in scale and investigate more sophisticated observers. They believe that constructor theory has the right tools to do this because information is also crucial in biological systems, which have, after all, generated human intelligence.

Living Systems

It’s long been a mystery how the microscopic laws of physics, which act on elementary particles without purpose, could give rise to life and evolution, which are extremely complex and seemingly purposeful. Darwin’s theory of evolution gives a qualitative answer—the DNA in cells codes for various traits in an organism, and natural selection explains how organisms with traits best adapted to their environment are more likely to proliferate, passing these useful traits on to successive generations. Constructor theory can help to elucidate this mechanism further, by framing the process in terms of the transfer of information, and how that information, in turn, becomes knowledge.

"Knowledge, in constructor theory, is defined as a particular type of information that can cause itself to remain embodied in physical systems," explains Marletto. This is Walker and Marletto’s starting point for developing what they call "causal mechanics." Their idea goes like this: Before life existed on Earth, the laws of physics could have triggered a chain of increasingly complex transformations, such as molecular reactions, carried out by increasingly advanced constructors. A simple cell, for instance, is a constructor that needs information in the form of DNA to carry out specific tasks. It also needs to be able to copy that knowledge and pass it on to new cells. "In this constructor-theoretic sense, knowledge is key to resiliency—a characteristic property of life," says Marletto.

Like cells, Walker notes, intelligent observers have the power to use information to cause transformations. The reason that traditional physics can’t account for this is that it is only concerned with micro states—the lowest levels of description you can think of—and what happens to them over time. At any instant, the configuration of atoms in your body is a micro state, while your body is a macro state. Physics assumes that the future of your body can be calculated by just evolving all those individual atoms forward in time.

But biology is far more complex than this simple picture allows. Atoms can group into molecules that only interact in specific ways with other molecules. And macro states seem to be able to influence individual particles. Take for instance animals that are able to regrow amputated limbs. This requires the animal’s brain to remember the location of the damage, store the information for some time and later act to make the cells in the local area produce a new limb. Similarly, the chain of reactions that allows you to simply raise your hand starts in your brain at a psychological level, causing specific electrons to flow in your muscles.

Emergence of Free Will

As evolution shows, biological systems are getting even more complex over time. Walker believes that is because this complicated interplay between different sub-systems using information opens up more possible states and trajectories for living systems to take than traditional physics allows for—potentially giving rise to emergent properties (S. Imari Walker & P. C. W. Davies, Journal of the Royal Society Interface, 10, 79 (2013)). Emergence is the phenomenon whereby a large collection of entities can display wholly new features that its individual components cannot. For instance, a gas has a temperature, based on the average motion of its particles: the faster they move, the hotter the gas. But the concept of temperature is meaningless if you try to apply it to any one of those gas particles individually. So temperature is an emergent property.

Emergence could explain how qualities such as specialized senses, intelligence and even consciousness evolved, says Walker. More controversially, she argues that nature can not only give rise to free will in this manner, it should favor the emergence of free will. Her line of thinking is that it may well be a principle of nature to create as many physical states and paths as possible. "If people can act as individual agents,"—with free will—"this will create more possibilities," says Walker.

But how do you actually show that this is how nature works? Walker intends to use computer models, known as cellular automata—simple grids of cells that evolve over time to form complex patterns according to simple rules like mathematical equations. Scientists use such systems to probe how complexity arises from simple systems. Walker will create a number of such cellular automata, each with a fixed and reversible rule (just like the laws of physics). Using the principles of constructor theory, she then wants to couple them together in such a way that each has the power to influence others only through their macro states. This means that the number of ways they can influence each other is increased, based on the possible number of macro states the systems can interact with. By contrast, in traditional cellular automata, the cells have less freedom to influence each other, being based only on their interaction with their neighbouring cells. Walker will then compare the results of her constructor theory model with those of traditional cellular automata, to work out which best represents living processes.

Sougato Bose, a quantum physicist at University College London, UK, believes constructor theory is a good candidate for becoming a meta theory in physics and says it might be useful as a basis for the cellular automata models. "Bringing these two together is not at all a bad idea," Bose says.

George Ellis, an FQXi member and expert on complex systems at the University of Cape Town, in South Africa, says that cellular automata aren’t always useful on their own for reaching broad conclusions. However, he says that Walker’s theoretical approach to understanding biological networks is "extremely interesting and important" and certainly opens up the possibility of incorporating free will into physics.