@Andrew Steane provided a really good answer. For further insight into his line of reasoning, I would recommend reading Chapter 1 of Information Theory, Evolution, and the Origin of Life (2009) by Hubert Yockey.

He writes:

The laws of physics and chemistry are much like the rules of a game such as football. The referees see to it that these laws are obeyed but that does not predict the winner of the Super Bowl. There is not enough information in the rules of the game to make that prediction. That is why we play the game. Chaitin (1985, 1987a) has examined the information content of the laws of physics y actually programming them. He finds the information content amazing small.

The reason that there are principles of biology that cannot be derived from the laws of physics and chemistry lies simply in the fact that the genetic information content of the genome for constructing even the simplest organisms is much larger than the information content of these laws (Yockey, 1992).

Bolded emphasis at the end is mine. While I have not read significantly further to really justify all of Yockey's claims, I think the spirit of his ideas is in the right place, and aligns with what Andrew as saying: just like you can't use the concept of a universal computer to specifically predict what programs will be written on it, you can't use physics to predict what the rules of biology will be. At best, you can only constrain the rules of biology to physically-possible scenarios.

If you choose a physics-dependent definition, then, yes, you can generally determine several biological rules -- but even then, not all of them, due to emergence, which others have mentioned. These often run the risk of being too specific -- we might have to expand the definition of life when we encounter new forms that we consider to be life.

However, I was merely noting that some people, including Yockey, have tried to produce relatively physics-agnostic definitions of life, which are of at least some epistemological interest (even if of little to no biological interest). I worry that Yockey's definition, which is based on information processing principles, runs the risk of being too general -- it is hard to see how computers, for example, would not be lumped into his definition of life. More generally, if several very different physical systems match a physics-agnostic definition of life, it is hard to argue that physics alone "predicts" how life ought to function: it simply constraints how life does/can function in our universe.

@Andrew Steane provided a really good answer. For further insight into his line of reasoning as I understand it, I would recommend reading Chapter 1 of Information Theory, Evolution, and the Origin of Life (2009) by Hubert Yockey.

Yockey writes:

The laws of physics and chemistry are much like the rules of a game such as football. The referees see to it that these laws are obeyed but that does not predict the winner of the Super Bowl. There is not enough information in the rules of the game to make that prediction. That is why we play the game. Chaitin (1985, 1987a) has examined the information content of the laws of physics by actually programming them. He finds the information content amazing small.

The reason that there are principles of biology that cannot be derived from the laws of physics and chemistry lies simply in the fact that the genetic information content of the genome for constructing even the simplest organisms is much larger than the information content of these laws (Yockey, 1992).

Bolded emphasis at the end is mine. While I have not read significantly further to really justify all of Yockey's claims, I think the spirit of his ideas is in the right place, and aligns with what Andrew was saying: just like you can't use the concept of a universal computer to specifically predict what programs will be written on it, you can't use physics to predict what the specific rules of biology will be. At best, you can only constrain the rules of biology to physically-possible scenarios.

If you choose a physics-dependent definition, then, yes, you can generally determine several biological rules -- but even then, not all of them, due to emergence, which others have mentioned. These physics-dependent definitions of life often run the risk of being too specific (e.g. too carbon chauvinist) -- we might have to expand the definition of life when we encounter new forms that we decide we want to be considered as alive.

However, I was merely noting that some people, including Yockey, have tried to produce relatively physics-agnostic definitions of life, which are of at least some epistemological interest (even if of little to no biological interest). I worry that Yockey's definition, which is based on information processing principles, runs the risk of being too general -- it is hard to see how computers, for example, would not be lumped into his definition of life. More generally, if several very different physical systems match a physics-agnostic definition of life, it is hard to argue that physics alone "predicts" how life ought to function: it simply constrains how life does/can function in our universe.

Edit: Let me clarify what I was saying in the above answer. (tl;dr: Everything depends on how you define life, and, in my opinion, the question of how best to define life is not a physics question, although it can be guided by a physics understanding.)

I'm going to use the following definitions for "physics", "biology", and "life".

• Physics: A question can be answered by physics, if it is of the form: "If X occurs, will Y happen?" In other words, physics discusses possible physical scenarios and processes, contingent on a set of physical constraints. Unless we're being speculative, the physical constraints match those of the real world.
• Biology: The field of biology answers questions, sometimes quite high-level ones, about the physical processes involving living things.
• Life: Now, this definition is up for discussion! (Which is my entire point.)

We can choose a physics-dependent definition of life -- e.g. "A living thing is a localized collection of carbon and other atoms with X additional properties" -- or we can attempt to produce a relatively physics-agnostic definition of life (which Yockey attempted to do -- the question of whether his definition is a good one is a related, but different, matter).

If you choose a physics-dependent definition, then, yes, you can generally determine several biological rules -- but even then, not all of them, due to emergence, which others have mentioned. These often run the risk of being too specific -- we might have to expand the definition of life when we encounter new forms that we consider to be life.

However, I was merely noting that some people, including Yockey, have tried to produce relatively physics-agnostic definitions of life, which are of at least some epistemological interest (even if of little to no biological interest). I worry that Yockey's definition, which is based on information processing principles, runs the risk of being too general -- it is hard to see how computers, for example, would not be lumped into his definition of life. More generally, if several very different physical systems match a physics-agnostic definition of life, it is hard to argue that physics alone "predicts" how life ought to function: it simply constraints how life does/can function in our universe.

But, in my view, exactly how you define "life" is not a physics question (it's a biology question), and because the set of allowed biological processes is strongly dependent on how you define life, I would hesitate to say that physics predicts "why the laws and behaviors observed in biology are as they are". As others have mentioned, even if you a priori choose a physics-dependent definition of life, as we already know on Earth, living systems adopt a wide variety of behaviors, none particularly dependent a priori on the laws of physics, but just as much as on circumstances, history, hysteresis, and emergence.