Sean Carroll is a physicist, cosmologist and research professor at the Department of Physics at the California Institute of Technology.
In 2010, Carroll was elected as a fellow of the American Physical Society for "contributions to a wide variety of subjects in cosmology, relativity, and quantum field theory and public science education".
This piece itself is based primarily on Carroll's book The Big Picture: On the Origins of Life, Meaning and the Universe Itself (2016). The chapters entitled 'Reality Emerges' and 'What Exists, and What Is Illusion?' are particularly relevant. It may be worth noting here that at various points in the following I found it hard to distinguish whether or not Sean Carroll was simply explaining/describing a theory/position or endorsing it. Perhaps that's because his primary purpose (at least in these instances) was educational.
It's the case that just about all scientists accept some kind of emergence. The kind they accept is usually classed as “weak emergence”. Indeed the philosopher Mark A. Bedau has said that “the notion of weak emergence is metaphysically benign”.
Physics (if we can speak in the singular) itself has its own definitions of emergence. Take this one:
"An emergent behavior of a physical system is a qualitative property that can only occur in the limit that the number of microscopic constituents tends to infinity."
In terms of examples. Biologists tend to see life or biology as an emergent property of chemistry. In chemistry itself it's believed that new properties are formed due to chemical reactions and suchlike. And in physics it's said that the properties of water (or even water itself) are emergent in the sense that they're unpredictable even after gaining knowledge of all the properties of its constituent atoms (of hydrogen and oxygen). In addition, in continuum mechanics we have the emergent properties of elasticity, viscosity, tensile strength, etc. Above and beyond all this are the examples which Sean Carroll himself cites in his book and elsewhere. Basically, there is a huge number of emergent properties in the natural world.
Strong emergence particularly often describes the whole's (or the system's) impact on its parts. That's sometimes categorised as “downward causation”. However, it can be said that there's no necessary link between strong emergence and downward causation, even if it strong emergence does occur.
The physicist Philip Anderson puts the case for strong emergence here:
“At each stage, entirely new laws, concepts and generalisations are necessary... Psychology is not applied biology, nor is biology applied chemistry.”
Murray Gell-Mann seems to strike a middle-way between strong and weak emergence in the following quote. Here's the hint at strong emergence:
“[I]t's essential to study biology at its own level, and likewise psychology, the social sciences, history, and so forth, because at each level you identify appropriate laws that apply at that level.”
And then Gell-Mann also hints at weak emergence:
“Even though in principle those laws can be derived from the level below plus a lot of additional information, the reasonable strategy is to build staircases between levels both from the bottom up (with explanation in terms of mechanisms) and from the top down (with the discovery of important empirical laws). All of these ideas belong to what I call the doctrine of 'emergence'.”
The philosopher Mark A. Bedau, on the other hand, puts the argument for suspicion of strong emergence when he writes:
“Although strong emergence is logically possible, it is uncomfortably like magic. How does an irreducible but supervenient downward causal power arise, since by definition it cannot be due to the aggregation of the micro-level potentialities? Such causal powers would be quite unlike anything within our scientific ken. This not only indicates how they will discomfort reasonable forms of materialism. Their mysteriousness will only heighten the traditional worry that emergence entails illegitimately getting something from nothing.”
Sean Carroll on Strong Emergence
Sean Carroll offers us us a definition of strong emergence. He tells us that
“[i]n strong emergence... [w]hen many parts come together to make a whole, in this view, not only should we be on the lookout for new knowledge in the form of better ways to describe the system, but we should contemplate new behaviour".
“In strong emergence, the behaviour of a system with many parts is not reducible to the aggregate behaviour of all those parts, even in principle.”
Yet if the emergent theory gives us “new knowledge”, and also shows us “new behaviour” (as Carroll puts it), then surely there must be new things (or new conditions) and therefore a new ontology too.
As an example of strong emergence, Carroll cites an example from condensed-matter physics:
“[C]ondensed-matter physicists have long argued that we should think of emergent properties as truly new, not 'merely' smeared-out versions of some deeper description.”
As for a specific example:
“Everyone working on the problem [of high-temperature superconductors] believe that such materials are made out of ordinary atoms, obeying ordinary microscopic rules; knowing that has been of essentially zero help in guiding us toward an understanding of why high-temperature superconductivity happens at all.”
Clearly the words “truly new” mean that it's not simply a new (or different) way of looking at the same thing: it is a truly new thing (or condition) which is being discussed.
Carroll also obliquely mentions “downward causation” when it comes to emergent systems. Or, less explicitly, he mentions the causal effects of a whole (or a system) on its parts. Carroll uses the example of a person to do so when he writes:
“A strong emergentist will say: No, you can't do that [predict the behaviour of an atom in the skin on the tip of a finger]. That atom is part of you, a person, and you don't [predict the behaviour of that atom without understanding something about the bigger person-system. Knowing about the atom and its surroundings is not enough.”
Elsewhere Carroll states that downward causation is what happens when the
“behavior of the parts is actually caused by the state of the whole, in a way not interpretable as due to the parts themselves”.
Carroll again describes downward causation1 when he says that
“there is an effect on that atom by the larger system of which it is a part – an effect that cannot be thought of as arising from all of the other atoms individually”.
We can accept that parts are affected by wholes/systems (or that wholes/ systems affect their parts). However, these wholes/systems may still be nothing over and above the sum of their parts. (Though here we're ignoring environmental or external effects on a whole or on a system.)2
Higher Level or Emergent Properties?
At many points when Carroll speaks of “emergent properties”3 one suspects that the words “higher-level properties” would work just as well. (Indeed Carroll himself interchanges these terms.) That is, perhaps it's the case that not all (or even any) higher-level properties need also be seen as emergent properties. Indeed, to a non-physicist, it may seem strange to class “temperature” and “density” as emergent properties (as Carroll does). Why not simply see them as simply higher-level properties? (This, admittedly, may be a difference which doesn't really make much of a difference.)
In one place Carroll cites hurricanes as an example (“just atoms in motion”). Not only that: he compares hurricanes to consciousness in that he concludes: “Why should we treat consciousness any differently?” However, isn't this a bad comparison? Surely a hurricane is perfect example of what's called “weak emergence”; whereas consciousness is often deemed to be a case of “strong emergence”.
Carroll makes a related point when he compares consciousness to water in terms of the latter's “phase transitions” (often also called “phase changes”). Firstly, we have water whose “molecular structure is  rearranged” (i.e., to form liquid, ice and vapor). And then Carroll says something similar about consciousness.
Perhaps the problem here is Carroll's stance on reductionism. Take Francis Crick when he describes mentality in an often-quoted passage:
“'You', your joys and your sorrows, your memories and your ambitions, your sense of personal identity and free will, are in fact no more than the behavior of a vast assembly of nerve cells and their associated molecules.”
Carroll too may misconstrue reductionism when he uses a phrase like “hurricanes are just atoms in motion” (which is like Crick's “are in fact no more than...”). (Hurricanes and water, in Carroll's accounts, can be seen as examples of weak emergence.) Elsewhere, he also says that we can think of ourselves as just “a collection of atoms”. The thing is, not many reductionists think of anything as just "a collection of atoms". They think of any x as a collection of certain types of atoms, forming certain kinds of molecules, forming certain kinds of configurations, forming certain kinds of systems which themselves exist in certain environments and interact with other systems... almost ad infinitum.
Carroll also asks us if
“there is something we learn from studying emergent level that we could not understand by studying the microscopic level”.
It can be said (at least at a prima facie level) that of course it must be believed that we can learn something new otherwise emergent theories would be deemed pointless. Thus, at least according emergentists, emergent theories must offer us something new - by definition. It's also the case the emergentists are bound to believe that “we'll learn more (and more quickly) by studying those higher levels themselves” (rather than “lower levels”).
Autonomous Emergent Theories?
“there is only one, unified, physical world, but many different ways of talking about it, each of which captures an element of reality”.
The obvious question to ask here is: How do we know if it's the same “element” described by a different theory or a different element entirely?
At a certain level, Carroll's position on emergent theories to some extent mirrors Ludwig Wittgenstein's doctrine of a “language game”. In Wittgenstein's case, language games are autonomous and play by their own rules (as it were). This, to Wittgenstein, had many important philosophical, moral and social consequences. The same appears to be the case with Carroll's position on emergent theories and his poetic naturalism.
Thus Carroll makes a lot of emergent theories being what he calls “autonomous”. He writes:
“The emergent theory is autonomous... it works by itself, without reference to other theories...”
Elsewhere, he says that with strong emergence “all stories are autonomous, even incompatible”.
However, in different places, Carroll also stresses emergent theories and their compatibility with non-emergent theories. Indeed Carroll hints at a lack of complete autonomy when he admits (if that's the right word) that “we might learn a little bit about higher levels by studying lower ones”.
It's nonetheless true that an emergent theory needn't explicitly (or even implicitly) refer to more fundamental/basic theories. However, surely it must still have various connections to them. In addition, if we refer back to Hedwig Wittgenstein's language games, why can't autonomous language games (even if they exist) include (or rely on) other language games. Indeed they do!
Carroll also emphasises the “mapping” of a fundamental theory onto an emergent theory in what he calls “course-graining”. Thus how can we have mapping as well as autonomy? Carroll gives the example of “an explicit map from one theory (molecules) to the other (fluid)”. (These are like the “bridge principles” - or “bridge laws” - formulated by philosophers.) Yet it's very clear that such course-graining (or bridge principles) are far less convincing or workable when it comes to mapping mental states to physical states then it is for Carroll's example of molecules and fluid. Indeed Carroll's example works from the fundamental theory (molecules) to the higher-level theory (fluid). Yet, in the philosophy of mind, philosophers have attempted to work the other way around: from mental states (higher-level states) to brain/neuronal states (lower-level states). That is, how easy is it also to map fluid (or states of fluid) to molecules (or systems of molecules) rather than the other way around?
So we must ask if an emergent theory can be truly autonomous. It can indeed be separate; though why also completely autonomous?
In addition, can “different ontologies” liv[e] happily alongside “the same underlying reality”, as Carroll argues? If not, then emergent theories, again, may not be completely autonomous. Indeed Carroll himself says that “emergence is about different theories speaking different languages”: not “deriving one theory from another”.
In a seminar Carroll also uses the word “consistence” in reference to the fit between emergent and more basic theories. Thus how can that consistency between two “autonomous” theories be established?
There are other problems here. Carroll says that an emergent theory can provide us with “an accurate description of the world within its domain of applicability”. Yet can an easier option be to simply say that it is the same domain which has different words, terms, concepts, senses and themes applied to it? This may simply depend, however, on how strongly (or widely) we take the word “domain” to be.
So who says the emergent theories are accurate? Does Carroll simply assume here an accuracy that's essentially guaranteed by the/a more fundamental/basic theory, thus limiting the emergent theory's supposed autonomy? Carroll also assumes compatibility between emergent theories and more fundamental (basic) theories.
Carroll talks about the ether theory. He says that it “served no empirical purpose”. That claim too assumes that a/the fundamental theory must have trumped the emergent theory. (That's if the ether theory was/is about emergence at all.) Surely both “empirical purpose” and accuracy can only be determined and guaranteed by fundamental/basic theories. Thus, again, the emergent theory can't be fully autonomous. (On a sidenote: the ether theory was deemed to “serve an empirical purpose” - at least at one point in scientific history.)
In addition, it simply isn't true that “different vocabularies” imply (or entail) “different ontologies”. At a very crude level, if someone uses the name the “Morning Star” and another person uses the name the “Evening Star”, and both persons know that both names refer to the planet Venus, then we don't have different ontologies on our hands here. Instead, different words, senses, technical terms, etc. can be seen to have the same ontology and therefore the “same underlying reality”. Ontology, after all, is often seen as what is, not as what we say about what is. Having said that, even an anti-realist can happily admit that different vocabularies don't necessarily imply (or entail) different ontologies.
There's also a problem with Carroll's assumption that if emergent theories can display what he calls a “useful role”, then that must automatically stop them from being “illusions”. Thus can we conclude (on Carroll's behalf) that consciousness, for example, plays a useful role and is therefore not an illusion?
What is a useful role anyway? And isn't it the case that on some interpretations illusions can indeed be useful? Can't “white lies”, for example, be useful?
So perhaps we can say that all this talk about “different stories” is really talk about about the practicality and simplicity (or the useful roles) of these emergent theories, rather than their meaty philosophical difference.
1) I recently noted an exotic case of downward causation in the films Split and Glass directed by M. Night Shyamalan. In Split, for example, a psychiatrist said that her patients had “changed their biochemistry with their thoughts”. The result? They turned into supermen and superbeasts.
2) The biologist Peter Corning wrote:
"The debate about whether or not the whole can be predicted from the properties of the parts misses the point. Wholes produce unique combined effects, but many of these effects may be co-determined by the context and the interactions between the whole and its environment(s)."
3) Any x can only truly emerge if at one point there was no x and then x emerged from something which was not itself x. So surely that can't be the case with Carroll's examples of water. (Perhaps not even with consciousness.) Instead, as soon as you have a particular set of conditions (say, H2O molecules), then you have x (say, water). This means that x doesn't emerge from those conditions because those conditions never exist without there being the property x. Of course we needn't see emergence in this temporal manner (i.e., not-x, then x); though it's hard to see it otherwise.