Why Information Grows

(the unpublished introduction)

NOTE: This is an introduction I wrote for Why Information Grows that I did not publish (I canned it around May 2014). I am now sharing it with anyone that wants to read it.

INTRODUCTION

People are not born with the ability to fly, cure diseases, or communicate at long distances, but they are born in a society that can endow them with those capacities. This book is about how these capacities emerge from our species ability to accumulate knowledge and information, and how the evolution of these capacities is limited by the need for knowledge and information to always be physically embodied.

In more poetic terms, this book describes the processes that allow our species to create what we imagine. It is a book about the networks people form. About the networks of bits and atoms that are begot by people, and about the networks of people that accumulate the knowledge required for us to transform the unthinkable into the mundane.

There are at least two ways of framing this book. The first involves posing a big question. Framing a book around a big question is appealing to readers who want to be clear at the outset about the book’s central focus and likely takeaways[1]. If you are looking for that defining question I will tell you up front that this book is about “why are some countries more prosperous than others?”

The premise I use to answer this question is that the process of economic development is the process by which society accumulates knowledge and information while battling the eternal requirement for both knowledge and information to be physically embodied. In simpler terms, I posit that the development of economies, just like that of any system, is the evolution of embodied knowledge and information, and that it is this embodiment what constraints the accumulation of both knowledge and information, and makes development difficult.

In the case of economies, information is embodied in products, like your home, refrigerator or car. These products are made primarily of information, since a victimless car crash does not destroy atoms, but the way in which these are arranged — and that’s information. Knowledge, in the most general definition I will adopt, is the ability of a system to process physically embodied information[2]. It is what you do when you write, or think, since both writing and thinking involve the processing of information that is physically embodied in pages of paper, files inside a computer or even your brain. Yet, when we understand that our world is made of physically embodied information we realize that most human tasks involve processing information, whether these tasks involve editing trash out of the city, translating a blue print into a house, or sorting socks after laundry. All of these tasks are processing physically embodied information, and although they are quite different, they follow our strict definition of knowledge. This capacity of processing physically embodied information, which we now know as knowledge, is embodied primarily in humans and in the networks that humans form, so its accumulation is constrained, among other things, by the social institutions that limit the ability of humans to form networks — such as family values and trust[3]. This book will be therefore dedicated to explain how information and knowledge are physically embodied and how this embodiment constraints the emergence of economic complexity.

It is worth noting at the outset that embodied knowledge and information do not evolve independently, since our species uses the products where we embody information to augment its capacity to accumulate knowledge[4]. Planes, email and Skype[5], allow us to weave networks that contain long distance relationships just like efficient shipping allows us to manufacture products using inputs sourced from far away lands. Yet, the networks that we build and the technologies we create do not remove the finiteness of our capacity to accumulate knowledge and information, and this finiteness is what forces us to quantize the knowledge and information that we accumulate into myriads of discrete products and bounded networks of people.

The discreteness of the “containers” that we use to accumulate knowledge and information is what makes accumulating both knowledge and information difficult. Ultimately, our society can only accumulate the volumes of knowledge that it is able to effectively subdivide and fit within the finite networks of people it is able to create. The same is true for information and products. As we progress, and develop institutions and technologies that allow us to weave increasingly larger networks, we develop the capacity to accumulate more knowledge, but the development of our capacity to build networks, and of our ability to embody knowledge is highly non-uniform in both space and time. That non-uniformity, and its dynamics, is what we call international differences in prosperity.

The finite knowledge carrying capacity of human networks transforms the process of economic development — that by which we accumulate embodied knowledge and information — from a continuous problem into a discontinuous puzzle. Over long periods of time people can solve this puzzle, albeit slowly, by growing the knowledge they have accumulated to include things that are similar to the ones they know how to do. Yet, since this process is slow, and highly path dependent, countries differ largely on the products — or packets of physically embodied information — they are able to make. Countries differ in the products they are able to make because they differ in the knowledge that is embodied in the human networks that they refer to as their nation and that produces these products. As a result, we live in a world where nations are no longer divided into bronze age and iron age civilizations, but into nations that differ in their ability to create modern products, from espresso machines to jet engines. Ultimately, differences in income are largely reflections of differences in the ability of countries to make products or provide services, which are technically differences in their ability to embody information in objects that people want to buy and knowledge in networks of people with capacities that others want to hire. So the necessity for information and knowledge to be physically embodied is what explains the ensnared diffusion of the knowledge needed to make products and it is what makes our world highly uneven.

Yet, there is an alternative way to frame this book, not around a question, but rather around a descriptive language. This is a framing that is not focused on answering “Why some countries are more prosperous than others?” but on the development of a language that integrates a number of disciplines and can be used to describe the complexity of our modern world in terms of matter, knowledge, networks and information. In that spirit, the book aims to describe the mechanisms that limit our society’s ability to pack knowledge and information into the physical and digital objects that we call products and that augment our capacities. If you like this framing, you will note that the book will answer the question: “Why are some countries more prosperous than others?” as as a corollary of the duality between matter and information that permeates the universe.

Thinking about contributions in terms of languages is useful because answers are, after all, nothing more than descriptions of phenomena. A scientist’s quest for answers is therefore not a constrained search, but an imaginative quest that includes the search for the best language that we can use to describe a natural or social system. Yet, a great descriptive language does not only allow people to describe what they observe, but also, it helps them penetrate the unknown and describe what they have not yet been able to observe. This ability to describe the potentially observable, to “fill in the gaps”, is what we often refer to as predictions, and it is what elevates a well-formulated description of reality into what some people call a theory.

Sometimes, to develop descriptive languages we need to hijack words. Often this requires us to embark on the abstract thinking required for mathematizing and narrowing down the definition of words that people already use in other contexts. We hijack words by weaving them with mathematical concepts and empirical observations, helping our language catch up with our understanding of nature. In some way, our search for descriptive languages or theories is a search for transformations that can turn the ineffable into the articulable. When the search is successful, we end up with a description of the world that can help us explain what happened, why it happened, and what were the actors and elements involved. Often these new actors are strange. Think of the many exotic words that our language has acquired during the last decades: genes, memes, proteins, quarks, photons, entropy and information, are just a few of these words. Often, the names of the actors that populate our descriptive languages narrow down the meaning of words that we were already familiar with: temperature, heat, force, mass and energy are just a few examples. In this book, I will use the words information, knowledge, networks, embodiment and complexity often. For information I will subscribe primarily to its physical definition, which is that of order and not necessarily communication — although to communicate we need to physically embody and transmit order. Knowledge, as I explained above, will be used to describe the ability of a system to process physically embodied information. Networks, will be used to refer to all systems that can be described as collection of nodes and links — from the power-grid to social systems — but often, although not always, will be used to refer to networks of humans. Embodiment will be used to highlight the necessity for information and knowledge to exist in a corporeal form. Finally, I will use complexity to talk about the holistic outcome of the systems I describe. In simple terms, a more complex economy would be one that has embodied a larger volume of knowledge and it is able to express it through the packets of physically embodied information it can produce (a.k.a products).

I bother to define these words because the answers to our questions are strongly constrained by our choice of language, and the meaning we give to words[6]. Problems that might be unsolvable, or difficult, using one way of describing the world, can become trivial when we cast them into the light of a different descriptive language[7]. That is why the power of framing a book around a descriptive language does not reside in its ability to answer questions, but in its ability to structure a way of thinking about the world that can help explore many questions[8], including the questions we use to validate our choice of language.

So you can probably guess why I am explicit about this second framing of the book: the framing where I explore a language that can be used to understand and describe the complexity of our world and the intimacy of the marriage between matter and information. This is because this language will facilitate our understanding of the process by which our species creates the objects that endow us with the ability to fly, purify water, play music, and communicate at long distances, but also, because this language will help facilitate our understanding of why the ability to create such objects is distributed unevenly across the world.

In the next pages I will build on concepts and ideas borrowed from information theory, network science, evolutionary biology, genetics, history, developmental biology, communication theory, organization theory, ecology, statistical physics, economics, sociology, economic sociology, political science and the arts. But why choose a language that borrows from such a diverse set of academic disciplines? My choice of language, while it might seem haphazard, is not the result of hedonistic choices, but the result of following the questions that I encountered while trying to understand the evolution of economic complexity. My effort to understand the information rich nature of the economy led me down the rabbit hole of information theory, whereas my effort to understand the ability of a system to pack and unpack knowledge and information has been inspired heavily by ideas from developmental biology. By the same token, the conspicuous expansion in complexity that characterize the economy, and that has been observed for eons in natural history, invited me to explore the non-human and non-biological origins of information, which are beautifully explained by far-from-equilibrium statistical physics. Similarly, my interest on humans networks lead me to explore theories of trust and social capital, just like my curiosity on the phenomena of economic development pushed me to study 20th century theories of economic growth.

At the outset, a book written from a combination of disciplines as diverse as economic sociology and statistical physics might seem overwhelming. Yet trust me, it is not. My contention is that the complexity of economies is such that attempting to describe it using only the language of a few disciplines, or a few related ones, is likely to result in loose ends; loose ends that we can often fail to see. The good news is that the language of most disciplines is now ripe enough for us to draw linkages between them that are deeper than analogies. Sometimes, these linkages help us realize that the exact same concept was developed by two different disciplines[9], and given different names and attributed to different authors. Sometimes, these linkages help us provide anchors that help solidify the foundations of high-level explanations. How good is an explanation of the evolution of the economy in terms of knowledge and information that obviates the basics of information theory? And how good is an explanation of the information rich nature of the economy that neglects the non-human origins of information described by the statistical physics of the late 20th century? The answer is probably not good enough. By the same token, how good is a description of the evolution and development of economies that fails to draw parallels and linkages with the evolution and development of biological organisms? And how good is an explanation of the ability of humans to accumulate knowledge in networks that obviates the basic social institutions that help humans form these networks in the first place? Alone, each scientific discipline is a candle in the dark. Together, these candles can begin to illuminate a path.

Finally, since the book does focus largely on the evolution of the world economy, I will make one final stop to discuss some of the distinctions that differentiate the approach I advance here, but also, some of the connections between this approach and those that are more commonly used to study economies. In the next pages I will highlight three important differences between my approach and the traditional mainstream[10] understanding of economies, and one crucial connection. Other differences and connections will be presented in the main text.

The three differences are:

First, I do not build primarily on assumptions of self-interest, or for that matter, on more realistic behavioral rules. Also, I do not deny the importance of the behavior of humans. Yet the decision-making aspect of humans is not crucial for the theory I advance in the next pages, since this is a theory based primarily in the finiteness of people’s knowledge carrying capacity instead of the infiniteness of their often assumed rationality or the quirkiness of their empirical behavior[11]. As is, I develop a theory that explains the behavior of economies based on the constraints that limit individual choices instead of the decision mechanisms that humans use to make these choices[12].

But why focus on the constraints of the options we face instead of the choices we make? When the options available to different countries, or individuals, are clearly different and evolve slowly — because they are constrained by the knowledge embodied in their networks — the force of the choices made by people, whether rational or otherwise, will be weak compared to the force imposed by the constraints that limit their choices. Think of a game of scrabble where one player has the letters A, E, S, T and R and the other has the letters Q, X, J, V and W. Certainly, the words that each player can build — if any — are quite different. As we will see, much like in the game of scrabble, we can explain much of the world’s differences in prosperity by looking at differences in the options available to countries, and the self-constrained evolution of these set of options, rather than their choices[13].

So for the purposes of this book, it will be enough to subscribe to a simpler model of humans, in which the finite capacity of humans to hold knowledge, rather than to make rational decisions, represents a fundamental source of economic order. In simple, caricaturist terms, I move away from a description in which humans are abstracted as calculators and explore the implications of a description in which humans are abstracted as knowledge carrying devices of finite capacity. A slow computer limited by the finiteness of its hard-drive, if you are willing to accept this machine metaphor as a clarification of my description of humans.

The second difference between my approach, and that of the economic mainstream, is the way in which I abstract products. By abstract, I mean the process by which we create internal representations of the physical and digital products that populate our world. Instead of focusing on the commercial value, or price of products[14], I focus on the capacity of products to embody information, carry the practical uses of human knowledge, and augment people’s capacities.

This difference speaks about the role of information in the economy. Prices, as Hayek forcefully and famously argued in 1945[15], are an information revealing mechanism that helps coordinate economic activities by revealing information about the supply and demand of goods. Prices, in Hayek’s interpretation, communicate information about the wants and needs of people and the availability of products and skills. The ability of prices to reveal information about supply and demand, and help coordinate economic activities, is unquestioned[16], albeit limited. Yet, my focus here is not on the information that is revealed by prices, which is information about a product, but on the information that is embodied in a product, which can be scarcely inferred from prices. This is the information that is embodied in the physical order of a product, the information that is destroyed when we demolish a building, and that we chew away as we eat apples. After all, all products, whether apples, oranges or lip balm, are all made of highly organized structures where bits are encoded in atoms[17]. Products are the offspring of the marriage between matter and information, information understood in the context of Shannon, Wiener, Turing and Prigogine, but not quite that of Hayek. This is the physically embodied information that separates cars from car wrecks and frogs from what’s left after your turned on the blender[18], not the information about supply and demand that is revealed in the price of a car or that of a glass of frog juice.

My third and final distinction is that macro structures, such as the economies of countries and cities, cannot be fully understood in aggregate terms — GDP, Population, Education Level, etc. — and hence, require the adoption of a more disaggregate language where types — what products and skills are present — rather than aggregates take center stage. This echoes the eternal cries of Wassily Leontief, the 1973 Nobel Prize winner who spent much of his life developing input-output models that could be used to describe economic structure. In his 1971 address to the American Economic Association Leontief wrote: “The time is past when the best that could be done with the large sets of variables was to reduce their number by averaging them out or what is essentially the same, combining them into broad aggregates; now we can manipulate complicated analytical systems without suppressing the identity of their elements.[19]

As Leontief insisted, and others echoed[20], we need to describe the components of the economy in a picture that allows for a finer resolution[21]. The language of networks allows us to achieve this higher resolution by helping us describe each component in terms of its connections with others. In same cases, this increase in resolution is simple. Instead of describing the economy of a country in terms of its population and income, we can describe it in terms of the products it is connected to. In other cases it is more nuanced, as describing products in terms of the diversity of the countries that export them[22].

These three differences will help us construct a description of the economy that is somehow different than the ones we usually hear about, but not completely. In fact, reinterpreting the economy as the process by which our species accumulates physically embodied knowledge and information will help us expand the economic growth theories that were advanced in the second half of the 20th century. Most of these are aggregate theories — connecting factors to aggregate output (GDP) — and are naturally constrained by their aggregate nature. Yet, we can nevertheless reinterpret and expand them in the light of the duality between matter and information that I describe in this book.

The models of economic growth advanced during the 20th century started by posing growth as the process by which economies accumulate physical capital — past production — relative to labor — population. Yet, the mismatch between the early theories of growth and the data available pushed economists to look for other factors to complete their explanation. The economic growth theories advanced during the last two decades of the twentieth century incorporated first, human capital — the knowledge of individuals — and then social capital[23] — the trust that enables humans to build networks[24]. Together, these three factors, physical capital, human capital and social capital, contributed enormously to our understanding of economic growth. Yet, they did not provide a complete picture. When we reinterpret these factors in the light of the duality between matter and information we realize that there is one factor missing, this is the knowledge that is accumulated inside the networks that humans form. The knowledge that is embodied in these networks is different from both, the knowledge accumulated by individuals — a.k.a. human capital — and the social “glue” required to form networks of humans — social capital. In the past, I have called this economic complexity, as the knowledge contained in these networks is expressed — albeit imperfectly — in the diversity and sophistication of the industries present in a location. This “economic complexity” is the difference between an army of aerospace engineers and NASA, the difference between computer science graduates and Silicon Valley, and the difference between a collection of soloist and an orchestra. So if you allow me to define this book yet again, this book is about the duality between matter, knowledge and information, and about how this duality implies the existence of a fifth factor of production. This fifth factor is not the physically embodied information that we call capital, nor the individual knowledge we call human capital. It is also, not the social glue that we call social capital, or the land from where we draw resources. It is not the knowledge that lives in individual humans, but in the networks they form. It is the knowledge and information that we accumulate collective thanks to our ability to forms networks of people and objects. So economic complexity is the heightened ability of our species to accumulate knowledge and information[25], and it is what endows our species with the unique ability to create what we imagine.

So now that we have reviewed the basic premise of the book, our choice of language, and the connection between our ideas and the traditional literature, it is time to conclude by summarizing the book’s outline.

The simple assumption that products are networks of atoms that embody the practical uses of human knowledge and imagination has profound implications. No man or woman has all of the knowledge necessary to create many of the products that we use every day, so our species is forced to quantize knowledge. This quantization is possible only when we can form networks that are large enough to hold the knowledge that we need to quantize. Products are useful for this, because by augmenting our capacities they enable us to accumulate more knowledge and information, and hence, create more complex products. This causes a co-evolution — a virtuous cycle — between the networks of atoms; we call products, and the networks of people that create these products. As it turns out, the ability of human networks to accumulate knowledge is what makes some societies flourish and holds others back, since knowledge is not only accumulated in networks of humans but also trapped in them. The requirement for knowledge to be physically embodied in networks of humans is a fundamental constraint, which is made even harder when the social capital needed to facilitate the formation of social and professional links is missing. So as I said earlier, the process of development is the process by which society accumulates knowledge and information while battling the eternal requirement for both knowledge and information to be physically embodied[26]. The need to embody knowledge in human networks is what keeps knowledge geographically circumscribed and help explain differences in the products that countries are able to make.

Part I of the book will be dedicated to the physical embodiment of information and will put a special emphasis on the origins and uses of physically embodied information. Part II will be dedicated to the networked embodiment of knowledge and will be focused on the constraint and factors that limit our ability to create the networks where we embody knowledge. Part III will put the lessons of part I and II in the context of traditional theories of economic growth and will explain how embodied knowledge and information explain differences in economic growth.

Our ability to create tangible instantiations of the objects conceived in our minds is a defining trait of our species, and is also, a defining property of human economies. Much of the information that we have accumulated during the last millennia comes from imagination. As the Nobel Prize wining biologist Francois Jacob reminded us[27]: “Whether mythic or scientific, the view of the world that man constructs is always largely a product of imagination.” Development is therefore, the battle to increase people’s ability to crystalize their ideas, since in this creative process is where we often find the solution to our species most pressing problems and where we can find some of the greatest sense of fulfillment[28]. Hence, the goal of development is not merely to include more people into the global webs of consumption, but to include more people into the global webs of creation. Our species has learned to organize into webs of humans that can create tangible instantiations of their dreams. Our goal is simple: to help each other construct the collective dream we call reality.

[1] Also, it appeals readers that like to read books by their cover.

[2] For other common uses of the word knowledge I instead use wisdom and experience. I define wisdom as the honing of knowledge that comes with experience, and experience as the adaptations that a system, or an individual, has acquired as a result of its interaction with outside stimuli. Experience, therefore, does not always lead to wisdom, since environmental conditions can reduce the capacity of an individual to process information, for instance, by inducing unnecessary fear.

[3] Chapter eight is dedicated primarily to the role of social institutions in the creation of large human networks.

[4] As a historical footnote it is worth noting that the idea of development as knowledge accumulation has a long history. In 1841 the economist Friedrich List wrote: “The present state of nations is the result of the accumulation of all discoveries, inventions, improvements, perfections and exertions of all generations which have lived before us: they form the intellectual capital of the present human race, and every separate nation is productive only in the proportion in which it has known how to appropriate those attainments of former generations and to increase them by its own acquirements.” (1841, p113)

[5] As I explain later, digital products are also always physically embodied.

[6] In The Language Instinct Steven Pinker also argues that, given the importance of language for understanding, conversations about semantics are not superficial, but profound.

[7] An excellent example here is the sum to fifteen game introduced by the magnificent 20th century polyglot Herbert Simon. This is a two-player game consisting on putting nine cards, from the ace to the nine. Players play in turns. Each player picks a card until one of them gets a combination of three cards that adds up to fifteen. The game is not super easy, and it takes a minute or so to complete. Yet, it is exactly the same game as tic-tac-toe, cast into a different language. (cite The Difference Scott Page)

[8] The powers of description are so ubiquitous that are often forgotten. Newton’s theory of motion, together with calculus, represents nothing more than a language that we can use to describe the movement of bodies. It does not tell us what mass is, or why it exists, but simply how mass affects motion. It does not tell us why motion can be predicted using second order differential equations; it just shows us how to use these equations to predict the trajectory of projectiles. Newton’s theory is not supported solely by logic, but by the fact that the predictions made by the theory work empirically[8]. Newton’s descriptions can help fill in the blanks of unknown motions, and by doing so, explain why the motion of projectiles follows parabolas, and why planetary orbits describe ellipses. The answers to these questions validate Newton’s theory of motion, but do not define it. Newton’s theory is not looking to answer a single big question, but to provide a language that can be used to answer questions of all sizes, including those he did not had a chance to think about. The same is true of most foundational theories. Foundational theories are descriptive. They provide the language in which questions are cast. Descriptions are therefore, the grammar defining “the box” where we think. Certainly, I do not claim to be making a contribution like that of Newton, which expanded “the box” substantially, but I hope his example helps illustrate that descriptive work is not necessarily trivial, and sometimes foundational.

[9] One of my favorite examples here is the development of the idea of cumulative advantage, Matthew Effect or preferential attachment. This idea was continuously rediscovered in different disciplines repeatedly during the twentieth century. Cumulative advantage is the idea that power-law distributions emerge in processes where increases are proportional to the share of the total that each entity posses. For instance, preferential attachment would be present in a world where each one of us has a different number of friends, and the number of new friends one acquires is proportional to the number of friends one has. This idea was described by Yule in the context of Darwin’s tree of life (1925), by Herbert Simon (who did cite Yule) in the context of biological distributions (1956), by Robert Merton as The Matthew Effect in the context of social recognition (1968), by Price in the context of the network of citations to scientific papers (1965), and by Barabasi and Albert in the context of network science (1999).

[10] I find it important to highlight that these are distinctions with respect to the mainstream of economics understood in broad terms. The field of economics is very critical of its own approaches and has a great capacity for internal criticism — although it has an awful ability to accept external criticism. So my claim is not that the distinctions I make have not been made before, but that they are not part of the most accepted view of economics, which in simple terms can be thought of as the economics that is taught to freshman and sophomore undergrads in college (see http://www.washingtonpost.com/blogs/wonkblog/wp/2013/11/30/colleges-are-teaching-economics-backwards/).

[11] Certainly, I acknowledge that during the second half of the twentieth century psychologist, and then behavioral economists, made great progress helping us understand the limitations of the overly simplistic model of human nature represented by the homo-economicus — an assumed being of perfect mathematical rationality. The contributions of psychologist and behavioral economists have not been simple critiques, however, but have also empirically mapped the richness of human behavior. The most famous of the scholars leading the departure from the traditional rational economic man are the psychologists and behavioral economists that use controlled experiments to study the quirks and irrationalities of human behavior. This is the school of thought started by the famous psychologists Daniel Kahneman and Amos Tversky — a school of thought that has shown that the foundational economic assumption that behavior is primarily rational — are not supported by empirical evidence. Yet, it is still worth noting that the empirical invalidity of the homo-economicus has not implied a complete departure of its use, especially by the most mathematically inclined economists that are fond of the mathematical simplifications that are brought by this oversimplified theory of human behavior”. For references see Kahneman, Thinking fast Thinking Slow, any of Dan Ariely’s book or the beginning chapters of Leonard Mlodinow’s Drunkard’s Walk.

[12] In this sense, my approach aligns more closely with some of the ideas that Thomas Schelling develops in Micromotives and Macrobehaviors. There, Schelling goes at length to explain that the macrobehaviors we observe in economies are often the result of constraints and conservations laws, rather than the emergent property of individual decision-making, since many equivalent micromotives, or micro rules, can give rise to the same exact macrobehavior.

[13] Recent additions to the theory of evolution can help explain my emphasis on the patterns that emerge from constraints rather than individual choices. In The Origin of Order, Stuart Kaufmann (cite) explains that there are two sources of biological order: natural selection, and self-organization. Kauffman argues that, when explaining biological order, modern interpretations of Darwin’s theory of evolution overemphasize the effect of natural selection and underemphasize the effects of self-organization. Kaufmann’s idea is that there is order in nature that emerges naturally and that evolution uses these ordered structures as building blocks. This is order that evolution gets for free, and that is essential to accelerate the evolutionary process — and move from a theory in which life is improbable, to one in which life is expected. For instance, to evolve an eye with spherical lenses evolution does not need to discover the sphere, since spheres emerge naturally from the physics of fluids. In the extreme case, in which self-organization is the dominant force, evolution acts more like a gardener, or a sculptor, instead of a designer, since it works primarily with an ensemble of forms that are provided by physical self-organization, such as spheres and fractals. Kaufmann’s argument is a bit more nuanced, and goes beyond the example of the physical spheres that predate the formation of the eye. His argument builds primarily on the dynamical states of random regulatory networks, which he uses to explain different cell types. Here, I take a position similar to Kaufman’s by claiming that, when it comes to economic variation, the force of individual — and mostly rational — choices has been overemphasized as the primary source of economic order. The main distinction that I am making here is that choices affect our world, but only within the space defined by option sets. This is a distinction that was often emphasized by Jane Jacobs (cite) who frequently remarked that material greed and self-interest can only operate over the ensemble of things that exist, and that therefore, the big question of economics are not just questions about the outcomes of human choices but about the mechanisms that give rise to the objects we choose from.

[14] Prices have been a central concern of economics since Aristotle’s distinguished between use value and exchange value of products in his first book of politics. Explaining prices was a major concern of the labor theory of value that permeated the writings of Adam Smith and Karl Marx, as well as of the marginalist theories of the late 19th century, which connect prices to the marginal utility of goods — the value people get from the last unit of consumption. When we focus on prices, products become numbers, numbers that helps detach products from their physical complexity and that are ideal for the study of commerce and finance. Prices represent the abstraction that is most adequate to understand the actions of traders, salesmen and bankers, since prices are a representation of products that focuses away from what a product does or is and focuses only on the profits that the people that own a product can obtain from it. Much of the study of economics during the last centuries can be seen as the exploration of the road that is opened by the route to abstraction that begins by replacing products by quantities and prices. During the last two centuries this route to abstraction has been explored, paved, and turned into a highway, while the road to abstraction that focus on what makes products useful — their ability to endow people with miraculous abilities — has been less explored. There is an important difference between the avenue that abstracts products as prices and the avenue that focuses on the ability of products to carry the practical uses of knowledge and endow people with miraculous capacities.

[15] Hayek, F. A. (1945). The Use of Knowledge in Society. The American Economic Review, 35(4), 519–530.

[16] Even though the cheerleaders of capitalism often overstate this point, weakening it by arguing that the price mechanism gives rise to a system that is almost perfectly meritocratic.

[17] Here, I am using the word atom in metaphorical terms to include all physical particles or fields, such as photon, electrons, etc.

[18] If in your imagination the frog was alive when you turned on the blender, that’s your problem :-).

[19] Leontief W. Theoretical assumptions and non-observed facts, The American Economic Review (1971)

[20] In The Competitive Advantage of Nations (Chapter 6 of On Competition) Michael Porter writes: “According to standard economic theory, factors of production — labor, land, natural resources, capital, infrastructure — will determine the flow of trade. A nation will export those goods that make most use of the factors with which it is relatively well endowed. This doctrine, whose origins date back to Adam Smith and David Ricardo and that is embedded in classical economics, is at best incomplete and at worst incorrect. […]

Contrary to conventional wisdom, simply having a general work force that is high school or even college educated represents no competitive advantage in modern international competition. To support competitive advantage a factor must be highly specialized to an industry’s particular needs — a scientific institute specialized in optics, a pool of venture capital to fund software companies. […]

Competitive advantage results from the presence of world-class institutions that first create specialized factors and then continually work to upgrade them.”

[21] As Leontief wrote in this 1965 article in Scientific American: “’Gross national product,’ ‘Total output,’ ‘Value added by manufacture,’ ‘Personal consumption expenditures,’ ‘Federal Government expenditures,’ ‘Exports’-these headings in the book of national accounts describe the familiar external features of the economic system. In recent years the students and the managers of the system have been confronted with many questions that cannot even be clearly posed in such aggregative terms. To answer them one must now look “under the hood” at the inside workings of the system.” Leontief W. The Structure of The US Economy, Scientific American (1965) Vol 212 No 4

[22] Since diversity is a network property — the number of products a country is connected to — the latter example is a case where the properties of second neighbors — friend-of-friends — is used to describe a node.

[23] When it comes to the development of social capital theory it would be unfair to attribute the credit purely to economists. Here, sociologists such as Ronald Burt, James Coleman and Mark Granovetter, and political scientists like Francis Fukuyama and Robert Putnam, deserve much of the credit.

[24] An alternative definition of social capital involves the opportunities to arbitrage information and resources that are open to an individual when he is connected to others that do are not connected to one another. This is what Ronald Burt calls the Social Capital of structural holes. This form of social capital, although important to explain differences of productivity between individuals, is not the one we are adopting here. Instead, we focus more on the social capital that is provided by dense networks (closed triads), which has also been discussed by Burt, but also by James Coleman (cite), Francis Fukuyama (cite) and Robert Putnam (cite).

[25] What anthropologists call cumulative culture. Richerson, Peter J., and Robert Boyd. Not by genes alone: How culture transformed human evolution. University of Chicago Press, 2008.

[26] You may ask: but what about politics, governance, and ethnic battles? Aren’t we forgetting much of the most common problems in developing countries? While these are all important constraints to the process of development, and we will discuss this later in the text, they are not fundamental in the sense that they can be modified. The finite knowledge carrying capacity of individuals and of networks of individuals, on the other hand, is here to stay. Governance can improve, ethnic groups can blend, but neither of these will remove the constraint that human capacities are finite. This makes the finite knowledge carrying capacity of humans, and of human networks, a fundamental constraint to the process of development that exists despite these other constraints.

[27] Francois Jacob (1977), Evolution and Tinkering, Science (1977)

[28] Certainly, there is a resemblance, but also a difference between this approach and that proposed by Amartya Sen’s in Development as Freedom (Sen, Amartya. Development as freedom. Oxford University Press, 1999). Development as Freedom focuses on the ability of economic development to remove constraint to people’s freedoms, putting special emphasis in political freedoms — such as the freedom to participate — and access to services — such as healthcare. The focus I present here is aligned, but instead of being focused on political freedoms and services, it focuses more on the liberation of people’s creative capacities.