1.3 Emergence & Epistemology

1.3 Emergence & Epistemology

In the last chapter I discussed metaphysics, which is concerned with what is and what can be, and I introduced complexity theory as a way of thinking about the world which allows you to describe systems in terms of their underlying causal mechanisms. I would recommend reading that first if you haven’t already. Lots of the ideas which relate to complexity theory will also carry over to the study of critical realism, which is the topic for this chapter. Critical realism is a philosophical system which is concerned with the philosophy of science, or epistemology. Essentially, what we can know and how we can know it. Are our senses reliable tools to navigate the world? Or do they represent only one small part of a much larger picture? Can reality be studied through materialist, scientific methods? Or does such a reductionist view of the world obfuscate the truth rather than reveal it? These are the sorts of questions we will be contending with today. But first, let’s pick up from where we left off in Wonderland. It was a thunderstorm, remember?

Part #1. Aunt Hillary

From your spot under the tree you watch as the storm slowly comes to an end and the skies begin to clear again. The sun comes out and you watch as the forest comes alive. You spot an ant trail on the ground next to you, trailing off into the woods, and you decide to follow it, hoping it might lead you to something interesting. You follow the trail as it connects back to a series of progressively larger and larger ones, until you are essentially following an ant highway, with hundreds of thousands of ants all marching in the same direction. They must be going back to the colony, you think. But instead of a hill, the parade of ants leads you to a little cottage tucked away in small clearing in the woods, with many ant trails coming from all sorts of different directions and congregating at this central location. Tentatively, you tip-toe over their criss-crossing paths and approach the door of the cottage. But before you can raise your hand to knock, the door swings open and you are greeted by a cheerful old woman. 

“Oh, hello dearie!” She says. “Come inside, I’ve been expecting you!” 

“I’m sorry, but who are you?” You ask. 

“Oh, just call me Aunt Hillary,” she replies with a grin, ushering you inside towards a seat by the fire and handing you a hot cup of tea. “I’m glad to see you got my message,” she says as you settle in. 

“Your message?” You ask. She shoots you a wink but says nothing. 

“What are all of those ants doing outside of your house?” You inquire. 

“Oh, I’m sorry! I thought you understood,” she says. “I’m Aunt Hillary, and this is the ant hill.”

“Oh! So they’re your ants?” 

“Well, only to the extent that the cells in your body are “your” cells. I didn’t choose to have them, nor do I have much control over them, but they make me who I am, yes.”

“I’m confused,” you say. “You’re not an ant.” 

“No, I’m not,” she replies. “No more than you are a collection of chemicals and cells, but that is what you’re made of, isn’t it?” 

“I guess I  hadn’t thought about it that way,” you say, taking a sip of your tea. “But how can you be made of ants if the ants are all outside while you’re sitting in here?”

“Oh child,” Aunt Hillary chides. “You forget that you are in Wonderland and here, looks can be deceiving. Here, look,” she says, gesturing to an open book. “Tell me what you see.”

“Looks like a dialogue of some sort,” you say as you examine the page. “Seems like the title of this chapter is ‘Ant Fugue’.”

“Quite right,” says Aunt Hillary. “A story written by a good friend of mine in fact, concerning this very topic. I can try to imitate some main themes of his for you now, if you like. You wish to know how I can exist independent from the ants which give me my form? Then first try considering where in this book the story resides.”

“Well, the words obviously.”

“Is that so? Not the letters?”

“Well, no, because the letters on their own don’t have any meaning. They have to be put into words in order to make sense.”

“And then the words are arranged into..?”

“...Sentences. Alright, I guess I see what you mean. It’s not just the words that matter, but the order in which they’re placed.”

“Precisely. Just as the same letters can be rearranged to create different words, the same words can be rearranged to create difference sentences. It’s not just the parts that matter, but the relationships between them. The words on their own don’t have any more meaning than the letters do, they are symbols. And those symbols must be arranged into sentences in order to convey an idea, which are then compiled into paragraphs and chapters to tell a story. But what happens if I, for instance, remove one adjective from once sentence in the text. Would it be the same story, or a different one?”

“Well, I can’t imagine what difference a single adjective would make. So I’m inclined to say it’s the same.”

“So then the ideas, to some extent, exist outside of the words which create them, do they not? If, for instance, a hundred people were asked to tell the story of Hansel and Gretel, I’m sure each version would be a little bit different, but the main plot point and characters would stay the same. Just like how a translated version of this book would carry the same contents while being composed of entirely different characters. The medium is not the message.”

“Oh, I get it,” you say. “The story somehow exists in a realm which is outside of and separate from the things which make up its telling. It’s not material, but conceptual.”

“Exactly,” says Aunt Hillary with an approving smile. “And you and I are no different. The letters on that page are like the neurons in your brain, or the ants in my colony. Taken in isolation they possess no intelligence or agency, they are simply firing or foraging, playing their small part in a much larger system. But just how letters come together to form words, neurons fire in clusters which activate symbols in your mind, and my ants operate in castes to perform different duties for the colony. However they are as unaware of me as I am of them. They may make us what we are, but what we are is more than what we are made of. Your subjective experience of being is not the firing of neurons, but the manipulation of symbols. At some point along the way, you, as an emergent entity, came to possess an agency which can act upon the very things which make it up. Aunt Hillary is just an idea, and yet in some ways I am more real than the ants themselves. I control their behaviour, but I don’t exist in a material sense any more than you do. You are not simply a series of electrochemical signals, otherwise what role would your consciousness play? You are an active agent. And that agency is not a product of your neurons or your nature or your nurture. It is something new, something more. So the reason why I can sit here and drink tea with you while my ants roam around outside is the same reason that you can sit here and drink tea with me, while all of your neurons are stuck inside of your brain. You are not in your head, yet that is where you came from. So, where are you? What are you?”

Throughout Aunt Hillary's speech you feel yourself overwhelmed by a wave of dizziness and dread. “I… I don’t feel so good,” you say, getting up abruptly from your spot by the fire and placing the teacup down on the table. In a distant daze you turn to walk out of the cabin and back, out, into the woods. 

Part #2. Critical Realism

The central idea I am trying to convey through this ant fugue, inspired by Douglas Hofstadter's dialogue of the same name, is that reality exists in stratified layers. Each one emerging out of, yet not being reducible to, the one that came before it. If you like, you can think of this in terms of different areas of inquiry. Mathematics leads to physics, physics to chemistry, chemistry to biology, and biology to psychology. At the most basic level, the firing of neurons in our brain is simply a collection of chemical computations, but that certainly isn’t what the subjective experience of consciousness feels like. We can use mathematics to describe neurology, but psychology is clearly a lot more complex than a series of equations. You cannot use mathematical models to predict psychological behaviour. Roy Bhaskar, the father of critical realism, described this phenomenon by saying, “it is true that the path of my pen does not violate any laws of physics, but it is not determined by them either.” So how can we construct a view of reality that accounts for this uncertainty?

Critical realism is a philosophical system that was designed by Roy Bhaskar to deal with the implications of emergence: both on what is, and on what we can know. In philosophy, these are the domains of ontology and epistemology. Bhaskar’s critique of modern science is that it prioritizes epistemology over ontology, emphasizing the ways in which we acquire knowledge, while overlooking the fact that there may be some limitations to what we can know through empirical methods. If reality contains emergent layers, then there is no reason to assume that all of it must be confined to the material realm. As Aunt Hillary demonstrates, the relationships between things can be just as, or even more important, than the things themselves. There may be aspects of reality which exist in a causal sense but not a material one.

In other words, critical realists believe in an objective reality, but they acknowledge the fact that our ability to acquire knowledge is constrained. What we can know is limited, relative, and often context-dependent. Although the knowledge we gain about the world does speak to real truths, they are approximations rather than absolutes. It’s like that parable about the blind men and the elephant, wherein a group of blind men encounter an elephant and—being unfamiliar with its form—each touch a different part of the animal, leading them to come to different conclusions about the whole. The blind man who touches a tusk is going to have a very different interpretation than the man who touches the trunk. Each one of their subjective experiences is correct, just not comprehensive. The information they perceive must be situated inside of a broader explanatory framework. What we observe represents only a small part of what is, and what is is only a tiny portion of what could be.

For Bhaskar, these are the realms of the real, the actual, and the empirical. “The empirical” is concerned with that which you directly perceive, the information provided to you through your senses or innovations in technology. The things you see, taste, touch, hear, or smell, as well as data collected through machines such as brain scans, mass spectrometers, or other material measures. “The actual” refers to that which exists, regardless of if it have been directly observed or not. So the ant colony, or your subjective experience of consciousness would fall into this category. A brain scan can’t tell you the thoughts running through your head any more than an inventory of ants can tell you anything about the colony that they are a part of. There are certain aspects of reality which may exist on the ontological level, but they cannot be observed through empirical methods. Finally, is the domain of “the real”, which is concerned with metaphysics and the casual mechanisms and consistent structures which generate events. This would be the complexity theory I talked about in the previous chapter. “The real” doesn’t care about who, what, when or where, only how and why. You can think of an iceberg as a useful metaphor. “The empirical” is the tip of the iceberg, which sticks out above the water and is easily observed. “The actual” is the rest of the iceberg, hidden underwater, out of sight and out of mind. And “the real” are the generative mechanisms which caused the iceberg to form in the first place. The physical, chemical, and mathematical characteristics which would cause any iceberg to form, not just the one you are currently observing. 

So, if the empirical is only one aspect of a much larger picture, how are we to gain insight into the nonmaterial, non-observable realms? Well, there is an important distinction in definitions I want to draw your attention to. There is a difference between empiricism and epistemology. Epistemology is concerned with what we know and how we can know it. Whereas empiricism is focused exclusively on knowledge obtained through sensory experience. Traditional scientific methodologies emphasize that which is measurable, material, and empirical. But this limited formulation leaves out the fact that valuable information can also be gained through our minds. It is not enough to merely experience things, they must also be integrated into a non-contradictory whole which requires moving beyond the empirical and towards the domains of the ontological and metaphysical. This is the key insight a critical realist approach provides: our sensory knowledge may be limited, but our minds still allow us to make meaningful inferences.

This is what is known as abductive, or retroductive reasoning. Unlike deduction, which goes from the general to the particular, or induction, which tries to go from the particular to the general, retroduction implies a regression, rather than a progression, in causal thinking. It is a form of inferential reasoning where events are explained by postulating and identifying the potential mechanisms which are capable of producing them. It’s how a detective pieces together clues at a crime scene. DNA is collected, interviews are conducted, testimonies are corroborated, and criminals are profiled. No one piece of evidence is enough to tell the whole story, a “big picture” interpretation is required to integrate all of the various clues most successfully. The goal is to develop an account of reality that carries the most explanatory power, but a number of different methodologies are available to draw upon. It’s not just the hard evidence that matters, and a lack of evidence doesn’t automatically mean someone is innocent. Motives and meaningful relationships between actors must also be considered. So if that’s the case, how does a detective know whodunnit?

This brings us to the final tenet of critical realism, which is “judgemental rationality”. This refers to the idea that although a variety of methods are available to acquire information, we do not need to embrace relativism in our assessment of reality. Our understanding of the world may be limited, but truth is objective. Therefore there must be some consistent criteria we can use to evaluate the likelihood of a given theory. To return to our detective example, a good theory is one which takes into account all of the available clues and information to provide a compelling chronicle of events. The best descriptions provide the most explanatory power while being able to withstand criticism and critique. For instance, if a piece of DNA evidence is found at the scene of a crime but the suspect has an airtight alibi, then there must be some alternative explanation as to how it got there. There may be conflicting clues, but that doesn’t mean that there isn’t one true series of events which caused all of them to come about. What we want is a theory that integrates all available evidence most persuasively, and these theories can be refined and developed over time as new information is acquired. The point is that we are capable of exercising rational judgement and being persuaded by the best argument. But this rational capacity is dependent upon what sort of evidence is provided to support a given claim, as well as the broader explanatory frameworks being appealed to to contextualize available information. Evidence alone is not enough, interpretation matters too. So what implications does a philosophy of critical realism have on how we acquire knowledge?

Part #3. Social Science

So, I’ve got a bone to pick with social sciences, and I’m going to take this opportunity to air my grievances. But first, what do I mean when I say “science”? I am referring specifically to information acquired through the scientific method, this being the standardized process of collecting and analyzing data to generate predictive results—usually by using experiments wherein variables can be controlled and manipulated to identify cause and effect relationships. Now, in the natural world, this process has worked quite well. Physicists, chemists, and biologists have all been able to use scientific experiments to gain meaningful knowledge about the natural world, and we have the scientific method to thank for much of the resulting progress we seen in the past few hundred years. So it makes sense that social theorists would try to replicate the success of science in the social realm. And replicate they have! Nowadays psychological, sociological, and political research is dominated by studies, correlations, and statistics. They make up the basis of psychology textbooks, social theories, and much of modern political debate. There’s only one problem… I don’t think it’s working. Let me give you some examples.

Starting with statistics, we’re all familiar with stats. These are numerical measures of some social variable, usually collected through census data or large scale surveys. People love throwing around statistics to make an argument. We’re all familiar with the numbers associated with things like the wage gap, wealth inequality, or police brutality. The only problem is that statistics alone don’t actually say anything meaningful about the phenomena they describe. They must be contextualized into a broader framework that allows you to make sense of them. For instance, the same piece of evidence—that women make $0.77 for every dollar a man makes—could be used to support feminist arguments demonstrating sexism in the work place, or an evolutionary psychologists position that men and women simply have different career interests. It doesn’t actually matter what the statistic says, it’s a matter of how it is framed. The fact that the top 1% of income earners hold 50% of the wealth is, in and of itself, a value-neutral statement. But if you start with the assumption that wealth should be normally distributed, it is going to seem like an indication of a broken system. However if you’re familiar with how complex systems tend towards 80/20 distributions, then you’ll know that income inequality actually means that the system is working as expected. The question remains as to if this is a desirable state of affairs, but these conclusions cannot be reached through statistical analysis alone.

The second model often employed by the social sciences is correlational research, where you take two data sets and analyze them to try to infer a causal relationship. As most people know, the problem with correlational research is that while any causal relationship can be demonstrated through correlational data, correlation alone does indicate causation. Even when two variables are related, this may be due to a third variable that has not been accounted for, or both variables may be indirect measures of the same underlying concept. Moreover, the vast majority of correlational research is used to confirm common sense associations rather than discover surprising ones. For instance, some study that demonstrates a positive correlation between reading and vocabulary size isn’t conducting groundbreaking research, it’s simply “scienceifying” something that I could have told you for free. This overemphasis on applying scientific methods to common sense understandings isn’t actually doing science, it’s just wasting time. The scientific method isn’t valuable because it confirms information we already know, its utility comes from its ability to reveal unexpected relationships. And this can only be achieved through the OG scientific method, which is experimentation.

So let’s talk about experiments. Hopefully we all know that experiments work by systematically controlling and manipulating variables in order to detect cause and effect relationships. This works extremely well in closed systems, which is what natural science tends to study, but completely falls apart when applied to social ones. Why is this the case? Well, it has to do with the objects of inquiry. As it turns out, people are a little different than rocks, rats, or nuclear reactors. We are conscious agents, and the content of our mind matters. Memories, interpretations, and expectations all play meaningful roles in determining our behaviour. You can’t conduct the same experiment on the same person twice, and you can’t make comparisons between subjects either. Controlling for external sociological factors doesn’t diminish psychological or biological ones, it amplifies them. Even if you were to come up with some unethical scenario where you take identical twins and raise them in a lab so you can conduct experiments without any potential confounds, the highly controlled nature of the study would make the results ungeneralizable to the general population. People are complex systems, meaning our behaviour is motivated by a plurality of factors that cannot easily be disentangled. 

What I am trying to get at is that there is a fundamental difference in kind which separates the social world from the natural realm. People are self-interpreting and value-oriented agents. Not only do a multiplicity of factors motivate any given decision, but people are often highly unaware of what these factors are or could be. Practicing social science requires operationalizing variables which are, by definition, subjective and context dependent. Unlike objective qualities like weight or temperature, there is no way to measure a concept such as happiness or anxiety, nevermind a metric that would allow you to compare subjective experiences across individuals. Unlike the natural sciences, wherein the objects of inquiry are independent from the aspirations of the researcher, social science both defines and influences social reality. Think about the rise in discussions about mental health over the past decade or two. The rise in conversations about mental health cause people to introspect, and potentially identify mental health issues in themselves, causing reports of mental health issues to increase and more people to be discussing it. It becomes a self perpetuating feedback loop.

Another major problem in social science is that it is impossible to isolate any phenomena down to a simple cause and effect relationship between two variables. Scientific experimentation relies upon “closed systems”, wherein one variable can be manipulated at a time while controlling for all others. Being able to exercise perfect control over passive agents is what allows the scientific method to produce predictive and replicable results. People, however, are much more complex than the objects of natural scientific study. Their actions are informed by a lifetime of experiences, expectations, interpretations, and biological mechanisms. It would be highly unethical to conduct a study which would attempt to control for all of these factors, and even then people have unique genetic predispositions which would only be augmented by controlling for all other variables. Given the near infinite amount of potential confounds at play, is becomes impossible to falsify a given claim. For a scientific theory to be valid it must be disprovable, however a theory tested through social research can always attribute failure to the existence of a confounding variable to justify an unfavourable result.

While many social scientists would readily admit that their findings are not nearly as precise, predictive, or objective as their natural scientific counterparts, few recognize that their research may actually be doing more harm than good. There seems to be a disconnect between the philosophy of social science, which acknowledges this fundamental difference in kind, and the practice of social science, which remains committed to a scientific ideal. The problem with social science is that it seeks to draw direct, 1:1 causal relationships between phenomena, rather than examining how a complex set of interactions lead to the emergence of social behaviour. There is never a single universal law at play, and any attempt to define such rules is only a partial truth, overlooking how a multiplicity of events are required to shape social action. Therefore any fact distilled from social research threatens to oversimplify and obscure the more nuanced aspects of social reality.

This is not to say that there are never any causal relationships that exist between social practices, but they are never all encompassing. They represent generalities or averages, rather than consistent absolutes. Personal beliefs and agency simply make up too big a part of the picture. For instance, take an issue like the use of corporal punishment on children, physically reprimanding them for bad behaviour. Nowadays it is generally accepted that this is not productive, bad for a child's psyche, and does much more harm than good. But there are still millions of people who were abused as children who went on to have successful lives, not because of, but in spite of the harm they endured. However there are people who would attribute their resilience directly due to a rough childhood. Although you can’t conduct a scientific experiment to prove that empirically, it makes sense intuitively. Acquiring the tools to overcome adversity makes us stronger and better at dealing with obstacles later in life. But clearly this principle doesn’t apply to everyone. It is entirely context dependent upon the child and their internal capacities to redirect struggles into success. This could depend upon genetic predispositions, social factors, or internal ones. Maybe reading a certain book is all it took to begin reclaiming and integrating your lived experiences. The point is that any number of factors could or could not play a role in shaping human behaviour. While one person may find success in spite of their abuse, another person might end up with a slew of mental health, attachment, and addiction issues. 

On the flip side, too much care and coddling can also result in adults that are not as well equipped to deal with the world as they grow older. Children raised by overprotective helicopter parents are also more at risk to experience anxiety and depression later on in life. A lack of personal responsibility, autonomy, and agency isn’t good for anyone either. You’ll notice that I may be citing facts derived through social research, certain correlations and relationships, but I am situating them inside of a broader explanatory framework. And I am not trying to argue that any one behaviour consistently leads to a given result. Alternatively, I am trying to demonstrate that different upbringings all along the social continuum, from extremely negligent to extremely overbearing, can have both positive, negative, and neutral outcomes. It always depends upon the individual, their life story and lived experiences. Certain factors may make a person more or less likely to result in a given outcome—fatherless homes are known to be correlated with criminal behaviour in adolescence. But there is always also going to be stories of people who gained more maturity and resolve for those exact same reasons. Individual agency and experience is always the main driving force behind any series of behaviours. You can never use one to derive the other, the relations are always going to be complex, entangled, and interrelated.

I also want to make it clear that when I talk about social sciences, I am specifically referring to research that seeks to demonstrate causal relationships between two socially defined variables. This leaves plenty of room for meaningful scientific research to be conducted in the realms of biology and neurology. Clearly we can conduct brain imaging research that correlates different brain areas with mental states, or test phenomena such as reaction time and memory. This sort of research is certainly more difficult because of the confounding social and psychological factors at play, but it is at least tethered to objective measures of reality. Cognitive psychology studies suggest that our working memory consists of 7 units, plus or minus two. But our minds are like a muscle, the more we practice a task the better at it we become. Waitresses who work without notepads probably have a better working memory than your average Joe, and there are master memorizers who can commit an entire deck of cards to memory within minutes. There may be norms, but there are also always meaningful exceptions. So there are no absolute laws at play like there are in mathematics, physics, chemistry or biology. An unwatered plant will die, I’ve heard that e=mc^2, and two plus two definitely make four. In the social world there are no such definite truths, the more abstract the concepts become the more room there is for uncertainty and interpretation. 

The emphasis on establishing facts rather than explanations reduces much of social discourse to who can provide the best empirical evidence to support their claims. However, as I alluded to earlier, this approach puts the empirical cart before the ontological horse, forgetting the need for a comprehensive theory wherein social knowledge is to be situated. The legitimacy of the facts produced by social science can always be called into question, an alternative explanation is always available, meaning factual findings hold little persuasive power over their theoretical opposition. Any statistic cited in social debate can be dismissed as inaccurate, incomplete, or inconsequential. For instance, in a discussion between the merits of capitalism versus communism, both sides would insist that a true version of their economic system has not yet been tried. Therefore they would not be persuaded by appeals to available evidence unless it supports their case. Ultimately, it is our values which inform our interpretation of the world, scientific evidence is only used as an anchor to legitimize our beliefs. Feelings don’t care about your facts. 

This is a position that was first introduced to me by philosopher Charles Taylor, and I believe it is of central importance. He proposed that the intelligibility of any argument relies upon shared value orientations. We must appeal to peoples emotional intuitions if we wish to persuade them to hold a certain point of view. For instance, I find it unlikely that anyone would support an economic system that leads to the mass starvation of millions of people. If communism or capitalism is more likely to result in that is a different question, but at least we have established a shared basis of understanding and discussion. We can investigate how we got to these starting principles later, but first we must recognize that we are being motivated by certain value orientations. People are much more likely to be persuaded by arguments that appeal to reason and shared ideals, rather than simply citing facts sourced through supposedly scientific studies.

This leads back to the critical realist notion that we should emphasize explanatory power over empirical evidence. The proof is in the pudding in terms of lived experience. If you can’t have a conversation with someone and meet them where they are at, recognizing their subjective interpretations and feelings, you’re not going to convince anyone of anything. People may be wrong in their underlying assumptions, but that is why it is so important to establish where you agree first as a scaffolding to build off of. And those starting points can come from anywhere, either appeals to empiricism, meaningful interpretation, or first principles. The point is that we want to focus on developing theories, not facts. A good theory integrates facts, but a collection of facts does not create a theory. In fact, a good theory can stand on its merits and predictive power alone, it doesn’t need to have any empirical evidence in order to be true. 

Historically, this has been the case. Social theory manifested and was communicated through stories and religion, wherein principles of human nature are expressed through narratives that convey morals and relationships between ideas. Our cultural heritage is full of folktales, adages and anecdotes. Although these stories may not be empirically true, they speak to deep social and psychological truths that we intuitively pick up on and understand. People are meaning and value-oriented agents, our minds haven’t evolved to processes the facts and figures derived through scientific research. Moral reasoning isn’t based off of calculations, but associations, and there’s nothing wrong with that. The ultimate goal of social inquiry is to develop ideas that allow us to better understand the world around us and how to act in a way which is purposeful and productive. This knowledge can arrive in any form, from philosophy to psychology, politics, or poetry. Domains based not just in ideas, but ideals. But where do those ideals come from? How can one make the leap from facts to values? How do our moral understandings and intuitions arise out of a material world? Well, that is what will be explored in the next chapter, where we dive into ethics, free will, and religion. 

Continue to Part 4: Free Will, Ethics, and Religion

---

Wonderland is a free publication that outlines a philosophy and political vision for the 22nd century. To support my work and receive updates on new posts, sign up for a free or paid subscription.