Abstract
Every scientific inquiry begins from something it treats as given: an animal, a market, a diagnosis, a culture, a measurement. This article argues that science starts late everywhere. Its starting points presuppose the material and symbolic stabilisations through which earlier coordination made them available, because science is itself an extreme case of poiesis, the coupled and cumulative deposition of coordination in durable material and symbolic forms. Its objects, from electrons to diagnoses, arrive already shaped by laboratories, archives, instruments, classifications, curricula and professional jurisdictions. The consequences of starting late vary with the kind of counterpart studied, and they become most distorting where starting objects and methods stabilised for counterparts that do not answer back are carried into living worlds whose counterparts do. The article applies the method of askability to the starting points themselves, asking what felt catch made each start pressing, which symbolic class escalated it into a question and then a discipline, and what the resulting stabilisation now conceals. It sets out twenty-one characteristic ways of starting late, groups them into five families, and identifies three general pathologies: concealment, selection and partition. Living Value Theory differs from other scientific frameworks in being a process ontology of how living coordination must work, grounded in five mediations irreducible to one another and with method following from it, where most frameworks are epistemologies or theories of particular processes. It offers a procedure for starting earlier and for reopening its own formulations, a test the article applies to the theory itself.
I. Every Inquiry Has Already Begun Before the Researcher Arrives
Every inquiry needs a starting point, and the fact seems too obvious to examine. A zoologist starts with an animal. An economist starts with a market, with scarcity or with choice. A psychologist starts with an individual, with behaviour, with cognition or with reported experience. A sociologist starts with institutions, groups or structures. An anthropologist may start with culture, difference, meaning or social relations. A physician starts with a complaint, a diagnosis or a pathological process. These starting points look practical. They seem to be the objects lying there, waiting to be studied.
This article treats the starting point as a variable. Disciplines largely decide what counts as given before any individual researcher begins, and the decision shapes what can then be seen, what can be asked and what will count as knowledge. A researcher who joins a discipline inherits its starting points together with its methods, and the starting points are harder to notice because they arrive looking like the world.
The central proposition is that most scientific research begins very late. Late means downstream, which is a different thing from false, illegitimate or useless. Late means: all the key assumptions about the nature of the object and the correct methodology of studying have already been made before any discipline or field is established. A diagnosis may be the right place for clinical intervention to begin, but countless processes had to occur before a diagnosis became possible: bodily disturbance, a felt complaint, a consultation, a classification of diseases, a profession entitled to apply it, an institution in which it is recorded. A price may be the right starting point for an economic calculation, but prices presuppose currencies, legal arrangements, commodities, infrastructures, practices of exchange and long histories. Culture, society, the individual, the human, language and species are sedimented objects in the same way. Each is the outcome of earlier coordination that made it stable enough to name.
The deeper hypothesis concerns the kind of being that does science. Humans are poietic animals. They coordinate through their bodies, with other beings and within the conditions of a place, as all living beings do, and they also deposit coordination in durable material and symbolic forms. Scientific inquiry therefore takes place inside worlds already densely shaped by multimateriality and multisymbolization. The problem goes further than the familiar point that researchers think through inherited concepts. Researchers inhabit inherited laboratories, universities, libraries, hospitals, archives, classification systems, measurement standards, databases, professional jurisdictions, journals and disciplinary institutions. Their starting points are poietically constructed starting worlds, built of matter as well as of ideas.
The method that allows these starting worlds to be examined is askability. Askability treats a question as an event with conditions of emergence. It reconstructs the felt catch from which a question arose, the symbolic class that had the resources to escalate it, and the stabilisation in which the question hardened into a doctrine, a category or a field. Applied to the history of ideas, it turns a history of answers into a history of questions. Applied here, it makes the starting points of the sciences themselves askable. Each starting point was once the answer to a question that someone had to ask, and the question has since been forgotten.
The article accordingly asks one question throughout. What happens when scientific inquiry mistakes the products of earlier poiesis for the starting structure of reality?
The scope needs stating at the outset. Science starts late everywhere, because all science is poietic. Electrons, stars, genes and chemical elements are scientific stabilisations with histories, instruments, classifications and conditions of access, and particle physics is as densely mediated by material and symbolic deposits as any human activity. What differs is the consequence of starting late. An electron does not interpret the category electron, and a star does not respond to the telescope as an interlocutor, so that abstraction and isolation can achieve excellent fit with such objects even though the objects reach the physicist only through an elaborate apparatus. LVT's ontological claims concern living coordination, and the physicist's practice is itself living coordination, but the objects of physics do not answer back. The difficulty examined here arises where starting objects and methods stabilised for counterparts that do not respond are carried into living worlds whose counterparts do, and where starting points inherited from human institutions are mistaken for the organisation of life.
II. Science Is Poiesis: Why Scientific Objects Arrive Already Made Available
Living coordination runs through five mediations. Three are zoetic: embodiment, being-with and dwelling, the coordination of a being with its own body, with other living beings and with the non-human-made conditions of a place. These three are sufficient for a complete living mesocosm. Poiesis is the capacity to deposit coordination into forms that outlast the interaction that produced them, and it generates two coupled mediations, multimateriality and multisymbolization, best understood together as one transformation with a material and a symbolic face. Multimateriality has long evolutionary precursors and many nonhuman elaborations: shells are secreted, nests and dams are built, shelters are carried, and some birds and primates make and combine tools. What becomes extraordinary in humans is the sustained coupling of multimateriality with multisymbolization, its cumulative scaling and its intergenerational sedimentation, so that deposits accumulate across generations and later coordination encounters them as ground. Science depends on exactly this. Plenty of animals alter matter; science becomes possible only where material and symbolic deposits are coupled, cumulative, portable, massively scalable and recursively available to later generations.
Science is an extreme form of poiesis. It is impossible without multisymbolization: terminologies, classifications, equations, hypotheses, diagnostic labels, formal models, graphs, measurements, citations, statistics, written protocols. It is equally impossible without multimateriality: laboratories, instruments, buildings, specimen collections, libraries, computers, scanners, microscopes, particle accelerators, clinical wards, archives, field equipment, recording devices. The two stabilise each other continuously. A scientific category becomes more durable because instruments and institutions enact it, and an instrument becomes meaningful because a symbolic system specifies what its output counts as.
Sleep research gives a compact example. Electroencephalography made it possible to record electrical activity from the scalp of a sleeping person, and in 1968 a standard manual set out how such recordings should be divided into stages (Rechtschaffen and Kales 1968). From then on, laboratories across the world could score sleep in the same way, and sleep stages became objects that could be counted, compared and treated. The physiological regularities are not created by the manual, but their stabilisation as universally comparable stages is a poietic achievement involving bodies, electrodes, amplifiers, paper and later software, scoring conventions, trained readers and institutions. What they do not tell anyone is what sleeping is for the sleeper and the people around her, and what has to hold in a household, a bed and a night for sleep to happen at all.
Classification systems show the same coupling at the scale of whole societies. Bowker and Star (1999) followed the International Classification of Diseases, the racial classification of apartheid South Africa and the classification of nursing work, and showed that each category was built into forms, files, databases and procedures until it became part of the environment in which people worked and lived. A category of that kind becomes very hard to see as a category, because it is enacted every day by the paperwork and the machinery through which it is applied. Researchers who later draw on the records inherit its divisions as data.
This is why disciplinary boundaries are so durable. Psychology, economics, medicine and anthropology are more than conceptual distinctions. Each has buildings, appointments, funding streams, databases, curricula, conferences, equipment, qualifications and journals. The division between them is poietically real without necessarily corresponding to any division in living coordination. Disciplines are very real achievements, and what is real about a discipline is not necessarily a natural joint in the phenomena it studies.
The distinction between human and animal shows how deep this goes. Humans remain animals completely at the zoetic level, and poiesis elaborates animal life without creating a second kind of life. Yet human and animal have become opposed categories, and enormous multimaterial institutions stabilise the opposition. Museums of natural history and museums of art, zoos and schools, faculties of veterinary medicine and of medicine, ethics committees for animal research and for human subjects, slaughterhouses and hospitals, departments of zoology and of anthropology: each reproduces a version of the distinction in buildings, rules and careers. A researcher who arrives in such a world encounters living beings already sorted into kinds, with different methods, different regulators and different moral standing. The sorting looks like a fact about the beings themselves.
The section thus arrives at a basic inversion. Science often mistakes the organisation of its access to an object for the organisation of the object itself.
III. The Hidden Career of a Starting Point: From Wonder to Discipline
Starting points do not arise arbitrarily, and they should not be criticised as though they had. Each has a career, and the career can be reconstructed with the method of askability.
The career begins with something that becomes askable. Successful coordination normally recedes from attention, and nobody needs to ask about what works. A question opens when coordination is disturbed, or when attention turns to something that works so well that nobody had asked how. The first opening is a catch at L2: a felt difference in ongoing coordination, before anything has been put into words, which may be negatively disruptive, positively arresting or simply newly salient. Negative snag and positive wonder are two forms of one structure. If a symbolic class, a group specialised in producing and managing symbolic stabilisations, has the resources and the reasons to take it up, the catch is articulated at L3 as a question. Repeated investigation then generates methods, instruments, categories and institutions, which stabilise the question at L4 as a field with a recognised object. The sequence runs from wonder or disturbance to question, to symbolic articulation, to material and institutional stabilisation, to a recognised object, to a method, to a discipline, and finally to a forgotten starting point.
At the beginning there are real questions. Why do prices vary? Why do organisms resemble one another? Why do people elsewhere live differently? Why do people fall ill? How does speech communicate? How can knowledge be justified? These questions are productive because something has become newly askable. Over generations, what began as an opening becomes a closure. The market becomes what economists study, culture what anthropologists study, cognition what cognitive scientists study and disease what medicine studies. The discipline inherits its predecessors' answers as the obvious ground from which new questions begin.
Three short careers show the pattern. In the early 1930s, as output and employment collapsed in the United States, the Depression greatly intensified the demand for systematic, standardised national accounting. Estimates of national income existed before, but they were irregular and not built to guide policy, and governments found it hard to say with any precision how large the collapse was. That catch was taken up by a symbolic class of statisticians and economists, and Simon Kuznets's report on national income, presented to the United States Senate in 1934, was a major institutional moment in the development that followed (Kuznets 1934). National accounts became an institution, and within a few decades "the economy", a single measurable whole with a rate of growth, had become the starting point of economic policy and of much economic research. What the starting point conceals is everything that sustains living coordination without passing through a market, which national accounts were never built to count.
The anthropological concept of culture has a comparable career. Edward Tylor's definition in 1871, of culture as the complex whole of knowledge, belief, art, morals, law, custom and other capabilities acquired by people as members of society, belonged to a Victorian comparative project deeply entangled with imperial encounters, in which the question was how to place the peoples Europeans met across an empire within a single history of humanity (Tylor 1871). Later anthropology transformed culture into a more localised ethnographic object, plural and bounded, with fieldwork as its characteristic access mode. What the starting point conceals is that people do not inhabit cultures in the way they inhabit bodies, households and places, and that the bounded group was a unit of comparison before it was a unit of life.
Psychiatric diagnosis has a third. The diagnostic manual published by the American Psychiatric Association in 1980 introduced explicit operational criteria for each disorder, answering a felt misalignment within psychiatry, the poor agreement between clinicians about what they were diagnosing (American Psychiatric Association 1980). The criteria were designed to improve agreement, that is reliability, which is a different matter from validity, since agreement about how to apply a category does not establish that the category carves psychopathology at its joints. They made diagnoses more reliable and made them countable, and diagnostic categories became the starting point for research, insurance, regulation and drug development. What the starting point conceals is the coordination of a life within which distress arose and in which it will or will not be remediated.
Each career ends in a starting point that looks like the world. Disciplinary starting points are therefore better treated as sedimented successes than as errors, and their strength creates their blindness. A discipline acquires a wonder profile: it becomes exceptionally good at seeing what its historical questions made salient, and correspondingly poor at noticing other dimensions of living coordination. Askability adds a further instrument here, since the absence of a question is evidence as much as its presence. A discipline's silences, the questions it never asks because its starting point leaves no room for them, map what its stabilisation conceals. When a discipline does turn to a neglected phenomenon, its starting point shapes the question it asks. Economists took up sleep in a well-known study of 1990, and the study treated sleep as an allocation of time, asking how the hours people sleep respond to wages and other demands on their time (Biddle and Hamermesh 1990). The question was a good one within its start. It could not ask what has to hold in a household for anyone to sleep at all, because the start had already made sleep into a quantity chosen by an individual.
IV. Twenty-One Ways of Starting Late
Table 1 sets out twenty-one characteristic starting points of the established human and life sciences. For each it states why the start comes late, what consequence follows, and what earlier question Living Value Theory asks in its place. The examples are broad strokes, since every discipline contains dissenters, and the concern is the default.
Table 1. Twenty-one ways of starting late
| Common starting point | Why it starts late | Characteristic consequence | Earlier LVT question |
|---|---|---|---|
| A kind of entity: animal, person, organisation | treats the achieved unity of the entity as given | requires later theories of interaction and relation | What coordination makes this appear as a coherent entity? |
| An institution: market, family, state, hospital | begins with a durable stabilisation | the institution explains instead of requiring explanation | What coordination has this institution stabilised? |
| A bounded group: culture, nation, community | takes boundaries as prior | inside and outside problems; cross-boundary consequences obscured | How are mesocosms actually joined? |
| A noun or category: depression, class, GDP, culture | begins with a symbolic compression | the process disappears behind its name | What process has the noun compressed? |
| A finished material object | starts after making has occurred | matter appears static; production and afterlife disappear | Through what relations to matter did this become possible? |
| One mediation | selects body, symbol, thing, place or relation | the other mediations return as "context" | Which mediations sustain this case together? |
| Very small or very large scales | jumps below or above lived coordination | the lived middle becomes an unexplained bridge | What happens in the mesocosm where life is coordinated? |
| Individual or structure | starts after a relational partition | requires reconciliations of agency and structure | What coordination produces both? |
| The deciding individual | privileges explicit choice | pre-deliberative coordination disappears | What happens before knowing, judging and choosing? |
| The human | makes specifically poietic capacities the paradigm for life as such | the zoetic ground of human life and its continuity with nonhuman life recede | What is shared across living coordination, and what does poiesis add? |
| The first-person subject | begins from what enters experience | smooth coordination and the coordination of others disappear | What works without becoming salient? |
| The theorist's standpoint | makes positionality foundational | accountability of knowledge can slide into treating the observer's position as constitutive of the object | To what coordination must the claim remain answerable? |
| The articulated | studies what became sayable | pre-articulate living disappears | What happened before words were required? |
| Meaning | begins downstream, in symbolisation | stakes beneath meaning are missed | What matters before it means? |
| Problems and breakdowns | samples what failure made salient | successful coordination vanishes | What works so well that nobody notices it? |
| Difference | makes contrast primary | common necessities of living recede | What must every case coordinate, and how does variation arise? |
| The measurable or exchangeable | begins with what enters symbolic equivalence | unrecorded value disappears | What sustains coordination whether or not it can be counted? |
| Cause or function | treats relations as externally specifiable | can suppress recursive responsiveness and stakes | What kind of counterpart is present, and what matters to it? |
| Truth or definition | begins once a proposition exists | the history of the question disappears | What made this question askable? |
| Isolation and control | makes phenomena tractable by subtraction | removed mediations may include the phenomenon | What was removed, and what did removal cost? |
| A single temporal scale | imposes one symbolic clock on processes with heterogeneous temporalities | synchronisation and desynchronisation appear as deviations from a supposedly homogeneous time | Which processes unfold at which rates, and how are they coordinated? |
The entries are manifestations of a smaller number of operations. They fall into five families.
Starting with things. The first family comprises the entity, the institution, the bounded group, the noun and the finished object. Its general problem is the reification of achieved stabilisations. The scientific object is encountered after a great deal of work has made it stable enough to name, and the work disappears into the name. Starting with separate things then generates a recurrent need to reconnect them. Much of the conceptual innovation of recent decades, relationality, entanglement, hybridity, assemblage, coproduction, consists of repair operations made necessary because the original units were treated as independent. The repairs are often brilliant, and they leave the units standing. LVT accepts that there are entities and treats the formation of an entity, the coordination through which something comes to hold together as one, as a question to be answered and not as a ground to begin from.
Starting with a privileged scale or mediation. The second family comprises one mediation, the very small or very large scale, and the individual or structure. Its general problem is premature partition. The body is studied without place, meaning without metabolism, materiality without being-with, the individual without the institution and structure without lived coordination. The omitted dimensions return later under labels such as context, environment, embodiment, embeddedness or interaction, and each return is presented as a discovery. LVT begins instead from concurrent mediation and asks, case by case, which mediation is doing what work, at which scale.
Starting with the exceptional capacities of humans. The third family comprises the deciding individual, the human, the first-person subject and the theorist's standpoint. Its general problem is taking poietic human capacities as the baseline of living. Explicit judgement, reflection and symbolisation become explanatory starting points although most living coordination proceeds without them, including most of human living. Humans are complete animals before they are symbolic specialists. They breathe, digest, sleep, move among other bodies and orient themselves in places through the same zoetic mediations as other animals, and their symbolic elaborations transform that coordination without replacing it. LVT therefore begins with continuities in living coordination and asks what poiesis subsequently adds.
Starting with what became salient. The fourth family comprises the articulated, meaning, problems, difference and the measurable. Its general problem is epistemic selection, since every member names something that has already become noticeable to someone. Sleep shows the consequence with unusual clarity. When sleeping works, little happens. Insomnia generates complaints, dreams generate stories, sleep disorders generate diagnoses, electroencephalography generates traces, and courts encounter sleep when something consequential happens during it. Research can therefore become extremely sophisticated about the symbolic and pathological edges of sleep while saying little about sleeping, the nightly coordination of bodies, beds, rooms, households and expectations that lets a person stop attending to the world and wake up in it. The counter-principle is one of LVT's most important: ask what works so well that nobody has noticed it.
Starting with the epistemic technologies of science. The fifth family comprises cause or function, truth or definition, isolation and control, and a single temporal scale. Its general problem is mistaking successful scientific operations for properties of the object. A controlled experiment can be exactly appropriate. Reduction is a tool whose value depends on what it is applied to. The questions that matter are what was removed, whether the removed relations belong to the phenomenon, and what kind of counterpart is being investigated. Those questions lead to the three pathologies, and then to the difference between the particle and the parent.
V. Concealment, Selection and Partition: Three Pathologies of Starting Late
The inventory condenses into three mechanisms.
Concealment. Successful poietic stabilisations hide their own production. A word presents a noun, an instrument presents a measurement, an institution presents a recognised object, and the researcher encounters the output without encountering the coordination that produced it. Symbols are well known to work this way: a working symbol presents its result and conceals the work that made the result available, and the concealment is part of what makes symbolic efficiency possible. The point generalises from symbols to whole research situations, which conceal their material construction in the same way. The sleep laboratory presents sleep stages, the clinic presents a diagnosis, the survey presents an attitude, and in each case the apparatus that made the object available withdraws as soon as it succeeds. Science forgets how its concepts were constructed, and it forgets, more completely, how its research situations were built.
Selection. The scientific record overrepresents what became salient. Problems, breakdowns, controversies, articulated meanings, measurable differences and institutional interventions all generate deposits: complaints, case files, lawsuits, surveys, statistics, publications. Smooth coordination generates almost none, because it prevents the disturbance that would have called attention to it. Misfit leaves an archive, and fit leaves very little. This is why the extraordinary can appear epistemically normal and the normal can appear extraordinary. A discipline that starts from its own record inherits the record's bias toward failure, conflict and the explicit, and it can mistake the shape of its archive for the shape of life. Legal history knows marriages largely through divorces, disputes over inheritance and prosecutions for bigamy, medical history knows bodies largely through illness, and the history of work knows factories largely through strikes, accidents and inspections. In each case the uneventful majority, the marriages that lasted, the bodies that stayed well, the shifts that passed without incident, left almost nothing for later researchers to start from.
Partition. A starting point produces a cut: body and mind, individual and society, organism and environment, nature and culture, human and animal, material and symbolic. Once institutionalised, each cut requires enormous theoretical effort to reconnect what it divided. The most rigid of these divisions can be called SWIPEs, split worlds with incommensurably paired epistemologies, in which a difference is reified into two independently constituted domains, each assigned its own way of knowing, with restricted transfer of method between them and a partition in place before inquiry begins. The most sophisticated repair theory can leave such a partition untouched, since it begins from the two sides it repairs. The division between the material and the symbolic shows the pattern within the human sciences themselves. Material culture studies and archaeology were given things, and the humanities and much of social theory were given meanings, so that a garment, a coin or a building has to be studied twice, once as an object and once as a sign, and then reassembled in a third literature on materiality. Every stage of the work is careful, and none of it questions the cut that made the reassembly necessary. Starting from coordination makes the cut visible as a cut, since a garment, a coin or a building is coordinated through matter and symbol at once before anyone separates them.
An important qualification follows. All research stabilises and selects, and inquiry cannot proceed without cutting. The question is whether inquiry remembers where it cut, how and why. Askability supplies the means of remembering, since it reconstructs the felt catch that a cut answered, the class that made it, and the stabilisation that has since hidden it.
VI. The Particle and the Parent: When Methods Travel Between Kinds of Counterpart
The strongest case for the established starting points is their success. Isolation, control, formalisation, repeatability and mathematical modelling have produced some of the most reliable knowledge human beings possess. They are especially powerful where engagement does not alter the object in the relevant respect. A particle does not change its behaviour because it knows it is being measured, a star does not adjust to the telescope, and a reaction does not reinterpret the chemist. For such objects, holding conditions constant and reducing the phenomenon to variables refines a symbolic relation to an object that has no coordination to lose. These objects too reach the scientist through apparatus and classification, and the difference lies in the fact that they do not answer back.
A particle and a parent are different kinds of counterpart. A parent responds to another person. A patient can respond to a diagnosis. A child changes because she knows she is being tested. An interviewee interprets the interviewer. A financial forecast can alter the market it predicts. A therapeutic interpretation becomes part of the relationship it describes. In each case the engagement of research enters the process it studies, so that the phenomenon on the far side of the method is partly made by the method.
A well-known case shows how much depends on recognising which counterpart is present. In the delay-of-gratification studies begun by Walter Mischel and colleagues at Stanford around 1970, young children could eat one treat at once or wait alone for a larger reward, and the time they waited came to be read as a measure of self-control, a capacity of the individual child. Later work changed the picture in two ways. Kidd, Palmeri and Aslin (2013) showed that children waited far longer when the adult running the study had just proved reliable by keeping an earlier promise, and far less when the same adult had proved unreliable. Waiting was shaped by the child's discernment of the adult, an interrecursive relation that the original design had treated as background. A large replication then found that the association between waiting time and later achievement was much smaller than first reported and was largely accounted for by family background and early cognitive ability (Watts, Duncan and Quan 2018). Waiting time turned out to be sensitive not only to an individual capacity but also to expectations about environmental reliability and to background conditions outside the laboratory.
The methodological mistake therefore lies elsewhere than in quantification, which can preserve its object perfectly well where the object survives isolation, as reflexes and eye movements do. The mistake is transporting procedures whose success depends on one recursive configuration into a situation with another, without accounting for the difference. The most widely cited case of that transport is the laboratory study of self-control, moral judgement or trust, in which what had made these phenomena what they are, the history of a relation, the stakes of a real situation, the presence of others who answer back, is removed before measurement begins. What is measured is real, and it is no longer what the study claims to describe.
The difference can be stated with three types of counterpart. A counterpart is nonrecursive when nothing about how it is engaged alters its course, as with a stone, a planet or a property boundary. It is selfrecursive when it responds to its own ongoing state, as a living body does when it regulates its temperature or adjusts to a drug. It is interrecursive when it is another being whose responses respond to one's own responses in turn, as in conversation, care, negotiation and teaching. Recursive type is independent of stability and predictability: weather is nonrecursive and volatile, a ritual greeting interrecursive and highly patterned. Real situations frequently mix all three, and a single entity can present different types in different aspects. A patient's blood chemistry, her monitoring of her own symptoms and her relation with her physician belong to one clinical encounter and call for different methods.
The laboratory is itself a poietic achievement designed in part to suppress unwanted recursion and to stabilise relations that behave as nonrecursive for the duration of the study. The randomised controlled trial, first used systematically in medicine in a trial of streptomycin for tuberculosis published in 1948, is a brilliant instance (Medical Research Council 1948). Randomisation and blinding manufacture comparability by suppressing the recursive effects of expectation and selection. That can be exactly what a research question requires. It can also remove what was supposed to be studied, where the effect of a treatment is inseparable from the relation in which it is given, and where the trial's own suppression of recursion makes the result describe a situation that exists only inside the trial.
The diagnostic question becomes simple to state and demanding to answer. What kind of counterpart does this method presuppose, and what kind is actually present?
VII. Starting Earlier: A Process Ontology, Not an Epistemology
An argument of this kind invites a misreading, according to which every other inquiry rests on assumptions while LVT sees reality directly. The misreading confuses two kinds of claim, and the difference between them is what most distinguishes LVT from other scientific frameworks.
Most frameworks in the human and life sciences are epistemologies or theories. An epistemology, whether positivist, interpretive, critical realist or standpoint-based, prescribes how knowledge should be produced and justified, and leaves the structure of what is known to the disciplines that apply it. A theory, whether rational choice, practice theory, attachment theory or niche construction, proposes how some particular process works and competes with rival theories of the same process. LVT belongs to neither kind: it is a process ontology, an account of how living coordination must work, of the structure that any living coordination has to have in order to be living coordination at all. Its claims specify that structure, within which theories of markets, attachment or ritual describe particular coordinations, so that they do not compete with those theories.
The claim of necessity rests above all on the irreducibility of the mediations. Any living being must coordinate with its own ongoing organisation, with other living beings that respond to it, and with conditions of a place that do not, and a being that has developed poiesis must also coordinate with durable material deposits and with durable symbolic ones. Each mediation answers to a different kind of counterpart, and none can be derived from another or substituted for it. A body cannot be coordinated with by coordinating with symbols, the responses of another being cannot be supplied by a place, and a found condition such as weather cannot be replaced by a made one, however much made things buffer it. Remove a mediation and the coordination it carried is lost, since no other mediation can take it over. There is accordingly no other way for living to proceed: three zoetic mediations for every living mesocosm, and five for every poietic one. The claim is open to a precise kind of refutation, a worked case of coordination that requires a sixth mediation or in which one of the five reduces to another, and it holds for as long as no such case appears.
Method follows from this ontology and does not precede it. Once it is clear that a counterpart answers back, methods built for counterparts that do not are ruled out for that aspect, whatever their prestige elsewhere. This is the decisive difference from frameworks that begin with a method and let the method define the object. It also clarifies what reopening means in LVT. The words living coordination, mesocosm, mediation, L1 and interrecursivity are symbolic stabilisations, and like any vocabulary they can harden, mislead and need revision, whereas what they name is claimed as necessary structure. LVT's proposal is therefore a discipline of starting earlier, with the structure every coordination shares, and of repeatedly reopening its own formulations of that structure, which is a different claim from an unmediated beginning.
Discernment comes before the architecture. The order of inquiry runs from the case, through discernment of what is vital in it, to its stakes, the mediations that carry them, the recursive structure of the counterparts involved, comparison with other cases and, only at the end, theoretical stabilisation. The reverse order, in which the architecture comes first and the case is sorted into its boxes, reproduces exactly the error the architecture was meant to diagnose. In inquiry, the five mediations, the levels and the counterpart types serve as aids to disciplined noticing, since the structure they name is present in every case and the question is how it is configured in this one, which is a different question from which box the case belongs in.
Starting earlier has an operational meaning. It means not deciding in advance that a case is fundamentally an individual, an institution, a culture, a disease, a representation or a material object. In place of that decision, the researcher asks a series of questions. What is being coordinated here? What must continue to hold? What has become salient, and what remains muted because it works? Which mediations are involved? What kinds of counterpart are present? What temporalities are unfolding, and at what rates? What had already been materially and symbolically stabilised before the researcher arrived? What has the method itself removed or added?
Askability turns these questions into a procedure for the starting points themselves. For any start a field takes as given, three steps follow. The first reconstructs the felt catch from which the start arose: what practical difficulty, recurrent failure, attraction or wonder first made this object worth isolating. The second identifies the symbolic class that escalated the catch into a question and the resources that allowed it to do so: which profession, with which instruments, which institutions and which vocabulary. The third traces the stabilisation and states what it conceals: what the start makes visible, what it leaves out of view, and what questions it has made unaskable. A fourth step follows from askability's treatment of absence. It maps the silences, the questions the field never asks, and reads them as evidence of where its start has drawn a boundary.
Sleep stages can serve as a short demonstration. The catch was clinical and scientific: recordings of sleep from different laboratories could not be compared, because each scored them differently. A symbolic class of sleep researchers, with electroencephalography as its instrument and the sleep laboratory as its institution, escalated that difficulty into a question of standardisation and answered it with a manual. The stabilisation made sleep comparable across the world and made sleep disorders diagnosable. It conceals the coordination that lets sleep happen in ordinary nights, the bed, the room, the people present, the trust that one can stop watching, since none of this appears on the recording. Its silences include almost everything about sleeping that never reaches a clinic. Once these steps have been taken, sleep stages remain an excellent access point for some questions and become visibly inadequate as a starting point for others.
A starting point that has been through these steps can still be used. An economist may begin with a price once she knows what a price is the stabilisation of, and a clinician may begin with a diagnosis once he knows what coordination the diagnosis compresses. The start becomes an access point, chosen for a purpose, and loses its standing as a foundation.
VIII. Science Starts from Knowledge: The Human Predicament of Scientific Inquiry
No other known animal has science in this massively cumulative form. Human scientific inquiry depends on an extraordinarily successful inheritance of material and symbolic deposits, and every researcher begins inside centuries of deposited coordination. That is science's great strength. It is also its characteristic epistemological danger.
The paradox can be stated in a sentence. Science begins with knowledge, which is why it can know so much and why it can fail to notice what its knowledge has already made invisible. The novice biologist does not rediscover species, since species arrive with the field guide, the museum drawer, the taxonomic code and the classification Linnaeus systematised in the eighteenth century (Linnaeus 1735). The medical student does not rediscover disease, which arrives with the textbook, the ward and the diagnostic manual. The economist does not discover the economy independently, since it arrives with national accounts and their categories. The anthropologist does not begin without culture, society, kinship, ritual, fieldwork and ethnography already surrounding the inquiry.
Science can accumulate because earlier coordination is deposited. Accumulation, however, converts earlier answers into later starting points. Every mature discipline therefore confronts a problem produced by its own success: its stabilisations become invisible because they work. The more successfully a field has stabilised its objects, the less reason its members have to ask how those objects became available, and the more completely its starting points disappear into the world it studies.
Training intensifies the predicament. Kuhn observed that scientific textbooks systematically rewrite the history of their fields, presenting current results as though they had always been the aim of inquiry and erasing the questions earlier researchers had actually asked (Kuhn 1962). Students learn the starting points as facts before they learn that anyone chose them, and a discipline reproduces itself by teaching its stabilisations as the world. The forgetting of starting points is therefore no accident of individual researchers. It is built into the way disciplines transmit themselves.
The predicament has a further consequence for the relations between disciplines. Because each discipline starts from its own stabilisations, the question of how their objects relate to one another has no place in any of them. Economics, law, medicine and the many anthropologies of particular practices each study institutionalised forms of living processes, and nothing in their starts says how those forms belong to the same lived world. Interdisciplinary work tries to build bridges between the starting points, which leaves the starting points in place. Starting earlier dissolves part of the problem, since institutions then appear as stabilisations within one mesocosm, related through the coordination they each stabilise.
The history of science has known this predicament in particular forms. Kuhn described how normal science works within a paradigm whose assumptions go unquestioned until anomalies accumulate (Kuhn 1962). His account concerns how communities respond to anomalies already identified as such. Askability adds the prior question of what made an anomaly feel pressing in the first place, and the analysis here adds a further one: what starting point made the anomaly appear as an anomaly, and what other phenomena that start keeps from appearing at all.
IX. Turning the Argument on LVT
The argument applies to LVT before it applies to anything else. A theory that has grown quickly across many domains has itself become sedimented ground. Its distinctions are increasingly available to its author and its readers without the work that produced each of them being recalled, which is exactly the condition this article diagnoses in other fields. LVT is entering the danger zone of disciplinary maturation.
The specific dangers can be named. The five mediations can become a classification scheme into which cases are sorted. The levels can become a ladder to be climbed. The counterpart types can harden into fixed properties of entities, when the theory itself holds that recursive type is attributed aspect by aspect. Mesocosmography can become a formula, and a growing archive can begin to determine which questions appear worth asking, so that recurrent readings of other thinkers produce predictable results because the template produces them.
Askability can be turned on the present question as well. The question of where research should start became askable within LVT through a catch that recurred across cases before it was put into words. The same pattern kept appearing: in sleep, which research knows mainly through insomnia and dreams; in anthropology's long concentration on culture and difference; in psychology's focus on articulated cognition; in the general preference for problems over coordination that works; and in the growing recognition that the disciplines are themselves poietic stabilisations. The recurrent question was why researchers keep seeing only what their disciplines have already made visible. It was escalated into an explicit question about starting points when the same pattern appeared from the other side, in the difficulty of explaining LVT to readers trained in established disciplines, who began from an entity, an institution, a meaning or a problem and met LVT's concepts as answers to questions they had never been led to ask. That account shows the argument's own conditions of emergence, which is what askability requires of every question it studies, and it leaves the argument's validity where it was.
The methodological injunction that follows applies to LVT's own deposits as much as to anyone's: do not mistake deposited poiesis for the organisation of life. Since no terms are immune from stabilisation, LVT can show its difference only by containing procedures for reopening its terms when cases cease to fit, and by using them.
X. Conclusion: Remember Where You Cut
Every inquiry has to start somewhere. An economist may appropriately start with a price, a clinician with a diagnosis, a psychologist with a reported experience, an anthropologist with an articulated difference. The error lies in allowing an access point to become an unquestioned foundation, and in forgetting that the access point was itself an answer to a question that someone once had to ask.
The argument can be summarised in five steps. Scientific disciplines are immense poietic achievements, and their multimaterial and multisymbolic stabilisations make cumulative knowledge possible. Science therefore starts late everywhere, since every discipline inherits entities, boundaries, saliences and research environments as though they were simply given. Starting late becomes especially distorting when methods and starting objects stabilised for one kind of counterpart are carried into recursively organised living worlds. Its recurrent distortions follow from concealment, selection and partition. LVT's alternative is a process ontology of how living coordination must work, through five mediations irreducible to one another and with method following from it, with askability as the means of making every starting point a question again and with its own formulations held open to reopening.
The resulting position lets the disciplines keep their tools and gives them back the memory of how those tools were made. Research cannot proceed without a cut, and the cut is where knowledge begins. The problem begins when the cut disappears.