Cited Papers
The published research behind the map, so a reader can follow any of our claims back to it. Two lists: the papers our framework itself leans on, written up in full, and every paper the walks cite across the map, each linked back to the systems that cite it.
Jump to: Papers our framework leans on · Papers cited across the map
Papers our framework leans on
An entry here is added only once we have read the paper in an open copy and checked the reference against the publisher's record; where the copy we read sits on a course page or a personal site, we name it without linking it.
On the roles of function and selection in evolving systems
What it argues. That evolution by selection for function is not special to biology: systems of many kinds, from minerals to living things, grow in complexity when many different configurations are tried and some are selected for what they do. The authors state this as a proposed law-like tendency, and they name the ingredients that make it work.
Why it is here. It is the closest published statement we know of to a claim our framework also makes: that selection for function is general across scales, not a property of life alone. Our approach is independent of it. We build from the engine's own record of walked systems, and we have not yet compared the two in detail. This entry marks the shared ground so a reader can judge the overlap for themselves.
Where we touch it. The Evolutionary Functions, Universal Information Flow in Geometric Space.
Time from quantum entanglement: An experimental illustration
What it argues. That the Page and Wootters idea, where a universe that is static as a whole still looks as if it is changing to anyone inside it, can be shown on a bench. The team entangled two photons and used one as a clock. Measured from inside (reading the clock photon and comparing), the other photon evolves; measured from outside (looking only at the pair as a whole), the state does not change at all.
Why it is here. It is the plainest demonstration we know of that "change" can belong to the view from inside a system while the whole stays still, which is the stance our ladder takes when it separates where a system sits (its rung) from when it arrived (the time pole). The authors call it an illustration, not a proof that our universe works this way, and we read it the same way.
Where we touch it. Universal Information Flow in Geometric Space.
Space from Hilbert space: Recovering geometry from bulk entanglement
What it argues. That space itself might be read off from how strongly the parts of a quantum state are entangled (linked so that they cannot be described separately). The authors take an abstract quantum state with no space built in, treat the amount of shared information between its parts as a distance, and recover a geometry from it, including how many dimensions it has. Small changes to the entanglement then bend that geometry in a way that resembles Einstein's equation for gravity.
Why it is here. It is a worked example of geometry coming out of information shared between parts, rather than information being laid onto a space that was already there, which is the direction the title of our framework paper points. The authors say plainly that the framework is incomplete: they assume how the state is split into parts and do not yet have time in the picture.
Where we touch it. Universal Information Flow in Geometric Space.
A Mathematical Theory of Communication
What it argues. That information can be measured. Any communication is broken into five parts (a source, a transmitter, a channel, a receiver and a destination), and the amount of information a source produces is a measure of how much choice there was in picking its message (its entropy). Every channel has a capacity, and as long as a source sends less than that capacity, the message can be coded so it arrives with as few errors as you like, even over a noisy channel; above capacity, errors cannot be coded away.
Why it is here. Our stacks are built on Shannon's picture: a signal leaves a source, rides a medium, and is caught by a receiver at the other end. He also set meaning aside on purpose, as irrelevant to the engineering problem, which is the line our model crosses when it adds floors for meaning and decision above the ones that carry the signal.
Where we touch it. Nested Stacks: The Universal OSI Model; Valence: One Force in Many Forms (where information is one of the things that crosses in an exchange).
Irreversibility and Heat Generation in the Computing Process
What it argues. That computing has a minimum physical cost. Any step that throws information away, such as resetting a bit to 1 whatever it was before, cannot be run backwards, and that logical one-way-ness must show up as a small amount of heat given off to the surroundings: about kT ln 2 per bit erased (k is Boltzmann's constant, T the temperature). The heat is not waste with no purpose: it is what makes signals standard and independent of their exact history.
Why it is here. It ties information to physics with a price tag: forgetting costs energy. That is the clearest published case we know of a gate's "rejection" step (dropping what doesn't go on) being a physical act with a measurable cost, not just bookkeeping.
Where we touch it. Nested Stacks: The Universal OSI Model.
More Is Different
What it argues. That knowing the fundamental laws does not let you rebuild the world from them. Large collections of parts show new properties that are not simple extensions of the parts, so each level of science (chemistry over physics, biology over chemistry, and up) needs its own new concepts and is as fundamental as the one below. His main tool is broken symmetry: a crystal, a magnet or a superconductor is less symmetrical than the laws that govern its atoms, and the whole ends up "not only more than but very different" from its parts.
Why it is here. It is the classic statement of the ladder: each rung obeys the rungs below it but is not just the rungs below it. That is the shape of our fold, where the same underlying events take on new properties once they are folded into a new kind of system.
Where we touch it. Universal Information Flow in Geometric Space (the ladder; the fold); The Evolutionary Functions (below the fold).
A General Model for the Origin of Allometric Scaling Laws in Biology
What it argues. Why so many features of living things change with body size by powers of one quarter: whole-body metabolism rises as mass to the three-quarters power, heart rate falls as mass to the minus one quarter, lifespan rises as mass to the one quarter. The authors derive this from the networks that deliver supplies through the body: if a branching network must reach every part of the body, if its smallest tubes (capillaries) are the same size in a mouse and a whale, and if it wastes as little energy as possible, the quarter powers follow. The same model fits blood vessels, airways, plant plumbing and insect breathing tubes.
Why it is here. It shows one distribution design, nested branching on a shared medium, producing the same numbers in animals, plants and insects whose tubes, fluids and pumps all differ. That is the kind of finding our stacks look for: the same shape, varied in form from system to system.
Where we touch it. Nested Stacks: The Universal OSI Model (nesting and routes on a medium).
Emergence of Scaling in Random Networks
What it argues. That many large networks (film actors who have worked together, links between web pages, the western US power grid, citations between papers) share one shape: a few nodes have huge numbers of links and most have few, with no typical size in between. Two ingredients are enough to produce it: the network keeps growing, and newcomers prefer to link to nodes that are already well linked. Take away either ingredient and the shape disappears. A side result: the oldest nodes keep pulling ahead of younger ones as the network grows.
Why it is here. It gives a simple, tested mechanism for a finding in our own record: credit for what later systems inherit concentrates in a few early first-doers. Our record arrives at that concentration through first-come ownership; growth with preferential attachment is a published mechanism that produces the same shape in other networks.
Where we touch it. Universal Information Flow in Geometric Space (credit concentrates in a few early owners); The Evolutionary Functions (ownership is first-come).
Threshold Models of Collective Behavior
What it argues. That when people decide whether to join something (a riot, a strike, a new idea), each has a threshold: how many others must join first before it is worth it to them. The outcome depends on the whole spread of thresholds, not the average. In his example, 100 people with thresholds 0, 1, 2 … 99 all end up rioting, one triggering the next; swap the single person with threshold 1 for one with threshold 2 and only one person riots. Two almost identical crowds give opposite results, so you cannot read individual attitudes off a group outcome.
Why it is here. It is a clean model of a gate that passes something on only once enough has arrived, and of a cascade that runs or stalls on a single gap. Both shapes come up in our record: thresholds at the gates of a stack, and cascades as one of the Attack Functions.
Where we touch it. Nested Stacks: The Universal OSI Model (the gates); The Evolutionary Functions (Cascade Induction).
Computation at the edge of chaos: Phase transitions and emergent computation
What it argues. That a system is best able to store, pass on and change information (the raw ingredients of computing) when it sits near the boundary between rigid order and chaos. Using cellular automata (grids of cells that update by simple local rules), Langton tunes a single setting and finds a transition like the one between solid and liquid; the most complex behaviour clusters near it. He goes further and proposes that life began near such a transition, and that evolution is life gaining control over more and more of the conditions that keep it there.
Why it is here. Its closing picture is close to our fold at life: below it, plain physics does the jobs when conditions allow (the done level); from it, the system holds its own state and keeps the job going (the held level). Langton states this as a hypothesis "if borne out", not a result, and we cite it as such.
Where we touch it. The Evolutionary Functions (below the fold); Universal Information Flow in Geometric Space (the fold).
Energy rate density as a complexity metric and evolutionary driver
What it argues. That one measurable quantity, the energy flowing through a system per second per gram of its mass, rises with complexity across all of cosmic history: lowest in galaxies, higher in stars, higher again in planets, then plants, animals, brains and human society. Chaisson proposes that energy flow and selection work together in every open system, living or not: energy drives systems away from balance, and those that can handle more flow per unit mass persist and grow more complex. He stresses that the individual values are rough estimates and that the trend is what matters.
Why it is here. It is a single number laid along the same ladder we use, from the early universe to civilisation, which makes it a natural yardstick to set beside our rungs. It is also common ground with our reading of Functions all the way down (2026-11-07): Chaisson extends a simplified physical selection to stars and galaxies, and our record books the same jobs at work there at the done level, with being held, selected and valued arriving at later folds.
Where we touch it. Universal Information Flow in Geometric Space (the ladder); The Evolutionary Functions (below the fold).
Statistical physics of self-replication
What it argues. That making a copy of yourself has a minimum heat cost set by physics. Starting from the fact that running a process backwards is always less likely than forwards by an amount tied to the heat given off, England derives a lower limit on the heat a replicator must release, set by how fast it grows, how much order it builds, and how long the copy lasts before it falls apart. A bacterium (E. coli) turns out to give off less than three times that minimum, and the bound also suggests why fragile RNA, not sturdy DNA, would have been the easier first self-copier.
Why it is here. It puts a physical floor under the most basic job of life, copying, and links it to the same kind of bound as Landauer's (England draws that link himself). He also notes that what counts as a "self" being copied depends on how an observer sorts the states, which bears on where we draw the line of the fold.
Where we touch it. The Evolutionary Functions (the fold at life); Universal Information Flow in Geometric Space.
Assembly theory explains and quantifies selection and evolution
What it argues. That an object can be described by how it could have been built: the fewest joining steps needed to make it from basic parts, reusing anything already made along the way (its "assembly index"). A complicated object that turns up in many identical copies is a sign that something selected it, because chance alone almost never builds the same many-step thing twice. The authors combine the two numbers (steps and copies) into one measure of how much selection a collection of objects needed, and propose it as a bridge between physics and biology, tested so far mainly on molecules.
Why it is here. It treats selection as something you can detect in the objects themselves, without first deciding what the "individual" is, which is the stance our walks take when they test a star and a cell with the same checklist. It also says the hardest step is making a thing for the first time, after which copies come cheap: the same shape as our finding that credit for a job flows down to whichever system did it first.
Where we touch it. The Evolutionary Functions; Universal Information Flow in Geometric Space.
Mineral evolution
Why it is here. It is a careful, rung-by-rung history of a non-living system getting more varied over time, and it says plainly that this is "evolution" only in the sense of change over time: minerals do not compete, mutate or pass anything on. That fits how our record reads Functions below the fold into life, at work at the done level but not yet held or selected, and it shows life reaching back down the ladder to reshape the rung beneath it.
Where we touch it. The Evolutionary Functions; Universal Information Flow in Geometric Space.
The Architecture of Complexity
What it argues. That complex systems of very different kinds (atoms, cells, organisms, firms, governments, problem-solving minds) are mostly built as hierarchies: wholes made of parts, which are themselves made of parts. His parable of two watchmakers shows why: the one who builds in stable sub-assemblies loses little when interrupted, while the one who builds a thousand parts in one go must keep starting over, so complex things built from stable intermediate forms arise far faster than things built any other way. He adds that such systems are "nearly decomposable": the parts inside each sub-assembly interact much more strongly than sub-assemblies interact with each other, which is why they can be understood one level at a time.
Why it is here. It is the classic argument for our ladder, where each rung is built from the rung below, and for nesting floors into floors: Simon's stable sub-assemblies are our systems, and his weaker links between them are our doors. He also notes the scheme needs no designer, only "survival of the fittest, i.e., of the stable", a reading of selection that holds below life.
Where we touch it. Universal Information Flow in Geometric Space; Nested Stacks: The Universal OSI Model.
Toward major evolutionary transitions theory 2.0
What it argues. That the big steps in the history of life (to cells, to the genetic code, to complex cells, to plants' energy factories, to many-celled bodies, to insect-style societies, to human groups with language) share a pattern: units that used to reproduce on their own come together and from then on can only reproduce as part of a larger whole, and new ways of storing and passing on information appear. The paper revises the 1995 list: it splits each transition into the forming, holding together and further change of the new level; it sorts transitions into "fraternal" (like units joining, as cells in a body) and "egalitarian" (unlike units joining, as a cell and the bacterium that became its mitochondrion); it adds small "limited" transitions that never spread widely; and it names an open question, new evolving systems arising inside an organism, such as the nervous system and the adaptive immune system.
Why it is here. It is the best-established account of systems climbing the ladder by joining into new wholes, and of the new information systems that come with each climb, which is the shape of our rungs above the fold. Its idea of "unlimited heredity" (a system whose possible variants vastly outnumber anything that will ever exist) is close to what we call a Seed.
Where we touch it. Universal Information Flow in Geometric Space; The Evolutionary Functions.
Life as we know it
What it argues. That any system with a boundary separating inside from outside (a "Markov blanket": the set of states through which the inside and the outside can only affect each other, as a cell membrane does) will, if it persists, look as if it is modelling its surroundings and acting to stay within its own limits. Friston gives a mathematical sketch and a simulated "primordial soup" in which such bounded, self-maintaining clumps form on their own. He notes that blankets nest (an animal's encloses its organs', which enclose its cells'), and asks what makes some blankets lifelike when there may be uncountably many.
Why it is here. It puts the boundary first: what a system is depends on where its inside meets its outside and what can cross. That is how our walks open (what is inside and outside a system), and how our doors work (the only places a flow can pass from one system to another).
Where we touch it. Nested Stacks: The Universal OSI Model; How We Walk a System.
Technological Approach to Mind Everywhere: An Experimentally-Grounded Framework for Understanding Diverse Bodies and Minds
What it argues. That every mind is a collective made of smaller agents (cells, tissues, organs), and that problem-solving comes in degrees rather than being simply present or absent. Levin proposes judging a system by the scale of the goals it can pursue (its "cognitive light cone": how far in space and time the things it works to change extend) and by what kind of intervention changes its behaviour best, from rewiring through training to argument. His main case is the body building and repairing itself, where cells coordinate through electrical signals to reach a target shape, as a form of thinking that came before brains.
Why it is here. It describes nested agents, each with its own goals and its own scale of concern, joined into larger agents through a shared medium (here, bioelectric signals between cells). That is close to our picture of valence climbing the ladder and becoming feeling. Levin himself uses "valence" for the self that collects rewards and punishments.
Where we touch it. Valence: One Force in Many Forms; Nested Stacks: The Universal OSI Model.
The Structure of Autocatalytic Sets: Evolvability, Enablement, and Emergence
What it argues. That an autocatalytic set (a network of chemical reactions in which every reaction is sped up by something the network itself makes, fed only by simple raw materials) usually breaks down into smaller sets that can each sustain themselves. These subsets can join, split and recombine, which gives the network something like inheritance, variation and competition, and one subset can make the very ingredient another needs to exist. The authors suggest sets of such sets could form a self-sustaining whole at a higher level, and, labelled as speculation, that cells, gut ecologies and economies might be read the same way.
Why it is here. It is Kauffman's line of "order for free" in its clearest open form: self-sustaining wholes that arise from enough mutual help, then help new wholes into being. The move from chemistry to ecologies and economies is the same cross-scale reading our framework makes.
Where we touch it. Universal Information Flow in Geometric Space; Valence: One Force in Many Forms.
Genetics and the Manipulation of Life: The Forgotten Factor of Context
What it argues. That genes, and organisms, are only understood in their context. Holdrege opens with a linden tree that grows tall and bare in a shaded forest and low and bushy on a sunny slope, and works through the history of genetics (Mendel, the chromosome, DNA, genetic engineering) to show where focusing on the isolated part is powerful and where its claims outrun it. Because an organism is a whole, changing one part changes the whole, so from the organism's point of view "there is no such thing as a side effect", only effects, some of them unwanted.
Why it is here. It argues for the habit our walks try to keep: read every part inside the whole it belongs to, and after any change look at everything the change touches. He also draws on Goethe's warning about the experimenter's inner enemies, among them haste, prejudice and the wish to prove a thesis directly, which we use as a check on our own method.
Where we touch it. How We Walk a System.
Each entry has its own link (this one is #wong-2023), so any page of ours can point to it. Our summaries are written in our own words and are not a substitute for the paper.
Papers cited across the map
Every research paper, review and preprint our walks cite, with the systems that cite it. Each system name opens that system's guided tour at Where we learned it. This list is rebuilt after every round of walks. Encyclopedia pages are linked on each system's own card and aren't listed here.
loading…