Sean Carroll: Teleology and the laws of science

Sean Carroll emergent features

“Teleology is absolutely acceptable in the laws of science,” says Caltech physicist Sean Carroll in this June 2013 talk (“Purpose and the Universe”) at the American Humanist Association annual conference.

Not the kind of statement one hears every day. Carroll begins by joking that they must have invited the wrong person, takes the keynote audience from quantum field theory to moral philosophy, and receives a standing ovation. His website is here and slides for this talk are here.

Starting at ~42:40:

We are very, very fortunate to live in a hierarchical reality, where there are ways to talk about the universe, other than using the fundamental laws of physics.

There are these vocabularies, these ontologies, these stories, these models — whatever you want to call them — that apply at these upper levels, and you don’t even need to know how the lower level works. And what’s interesting about this is, not only can you have different levels of description, but they can sound very, very different. Things that you thought were really important at the bottom level might not even show up at the upper level.

I’ll give you one obvious example — my favorite example of what we call an emergent feature in a higher-level theory: the arrow of time. The fact, we already mentioned, that the past is different from the future. The past is done. The future is up for grabs. I can make choices about the future. I remember yesterday; I don’t remember tomorrow.

Well, if you go back to Newton’s equations — if you are Laplace or you’re Schrödinger or whatever — that distinction between past and future is nowhere to be found. In the fundamental laws of physics, the past and future are on exactly equivalent grounds. There is no difference between the past and future as far as the fundamental laws of physics are concerned.

Yet, you would be crazy to try to describe human behavior, or biological evolution, or even the evolution of cosmology and the whole universe, without taking into account that in the macroscopic realm there is a very strong arrow of time. There’s a very noticeable difference between the past and the future.

So, it’s not true that just because something is invisible at the lower level, it can’t be there at the higher level. They just need to be compatible.

How do you make the arrow of time in the macroscopic world compatible with the reversibility of the microscopic world? It’s because of the big bang. Because in the past, 13.8 billion years ago, the universe started in a very, very special, delicately arranged state. It’s like a little wind-up toy that started there and has been puttering along ever since. Winding down. Dispersing its energy, as entropy goes up.

We have a match. We have a way of mapping the fundamental laws of physics and the macroscopic ones. It just requires knowing which macroscopic configurations are allowed. And you find that words — like: cause and effect, memory, choice — appear and are very important at the macroscopic level, even though they are nowhere to be found at the lower level.

So, I would say — having gone through this whole journey — that teleology is absolutely acceptable in the laws of science. It is on the table as a possibility. Teleloogy is the idea that the way to talk about a system is to say: It has a goal. It is trying to find something.

At the equation level of the fundamental laws of physics, there is no teleology. But that’s not the end of the story. At the higher level, in the emergent level, in the macroscopic realm, there might very well be a teleology.

Remember, you agreed with me when I said the Roomba has a teleology. It was built in order to vacuum the room. There was a purpose. There was a reason why somebody built that Roomba. And that person who built the Roomba is made of atoms, obeying the fundamental laws of physics.

Is there purpose in the universe? Of course. I have purposes. You have purposes. You came here, and you’re just made of atoms. It’s because you’re describing yourself at a different level of complexity.

So, if teleological language — finding a purpose, moving toward a goal — is the best scientific theory applicable to a certain level, then there’s every reason to talk that way.

  • VN Alexander 12 Jul 2013, 9:14 am

    I’m glad to see this post. I’ve claimed to be a “card carrying” teleologist for years. But I would not say that “Teleloogy is the idea that the way to talk about a system is to say: It has a goal. It is trying to find something.” Purposeful actions are made possible because living systems change the LIKELIHOOD of certain physicals event occurring over other physical events occurring. Such actions are not “other than the fundamental laws of physics.” When considering thermodynamic process it is the NUMBER of different kinds of outcomes that matters. With teleological processes, the FORMAL aspects of the molecules/things involved may affect the likelihood of certain kinds of reactions taking place over others, even if these formal properties are rare in the system. This leads to differentiation, and if this process tends to be self-reinforcing, it is evidence of FINALITY. I like my Aristotle without the Christian undertones.

    • Howard Silverman 12 Jul 2013, 2:27 pm

      Hi Victoria, thanks for your response and for the precision of your language. Reading this short paragraph makes me want to attend a biosemiotics conference.

      To be sure, Sean takes some shortcuts in this talk for (what I guess to be) a fairly general audience. He makes no clear distinction between goal-seeking and purposeful behavior, one that systems folks like Churchman and Ackoff have emphasized.

      Still, his description of teleology, like yours, does hinge on contingency. Perhaps because he speaks in terms of individual organisms and artifacts, whereas you stick with processes, different types of associations arise.

      In any event, thanks again. I enjoyed the book excerpt on your website.

Leave a Comment