Abstract
Living Value Theory holds that life is continuous recursive mediation in a mesocosm. This article asks what the word continuous can mean, and answers that it cannot mean homogeneous. Every mediation coordinates processes running at radically different rates, and continuity is the ongoing achievement of workable relations among them rather than a background container within which mediation occurs. Temporality, on this account, is relational structure among unfolding processes: order, phase, rate relation, nesting, anticipation, recurrence and interruption. Duration is one element of that structure and not the whole of it. Three consequences follow. The five mediations do not have temporalities but temporal bandwidths, and those bandwidths are grossly unequal, with the three zoetic mediations bounded above by the lifetime and dwelling outranging them by many orders of magnitude. The two poietic mediations extend, transform and transport temporal coordination beyond the limits of individual embodied enactment, multimateriality through durable material organization and multisymbolization through replicable symbolic stabilization. And the recursivity levels are not durations: the distinction between L1 and L2 is not between a longer and a shorter process but between coordination proceeding transparently and a difference within its own unfolding becoming felt.
I. What Does Continuity Require?
Living Value Theory begins from a single proposition: life is continuous recursive mediation in a mesocosm. Everything else in the architecture is a consequence of asking that proposition concrete questions. This article asks one that has been deferred. What does continuous mean?
The word carries a temptation. Continuity suggests a smooth flow, an unbroken passage, one thing following another at an even rate. Applied to living coordination it suggests that mediation happens within time, that time is the medium through which coordination passes, and that continuity names the absence of gaps in that passage.
This picture is wrong in a way that matters for the whole architecture. Coordination does not occur within a temporal medium. Consider what a single ordinary moment of coordination actually holds together. A person walking across a room is coordinating sensorimotor adjustments occurring over tens of milliseconds, a respiratory cycle of several seconds, a circadian state built over the preceding sixteen hours, a musculoskeletal condition assembled over decades, a relationship with the person they are walking toward that may be forty years old, a floor laid a century ago, a building code stabilized in the 1960s, a language whose grammar took millennia to settle, and a gravitational regularity older than the planet. None of these runs at the rate of any other. Several differ by more than ten orders of magnitude.
Continuity, in that moment, is not the smooth passage of a single flow. It is the fact that all of this holds together well enough for the walking to proceed without any of it requiring attention. That is an achievement, and it is a temporal achievement of a specific kind: not the maintenance of one rate but the maintenance of workable relations among many.
Differential temporality is therefore not an additional dimension of continuous mediation, arriving after the mediations and the levels have been set out and adding a further variable to each. It is what continuity consists in. Living mediation is continuous precisely because processes at radically different rates are held in relations that work, and it ceases to be continuous when those relations fail. There is no homogeneous temporal background underneath the mesocosm within which the mediations then unfold at their various speeds. The relations among the rates are the mesocosm's temporal structure, and there is nothing further underneath them.
This commits the article to a specific account of what temporality is, and the commitment should be stated at the outset rather than allowed to emerge. Temporality is relational structure among unfolding processes: order, phase, rate relation, nesting, anticipation, recurrence and interruption. Duration is one element of that structure. It is the element most easily measured and for that reason the one most often mistaken for the whole. Wherever this article speaks of rates and durations, those are components of a relational structure and not the substance of temporality itself.
Three consequences follow, and the article develops them in turn. The mediations must be recharacterized as bandwidths. The poietic mediations must be recognized as extending temporal coordination beyond the limits of embodied enactment. And the recursivity levels must be understood as bounded by something other than duration.
II. Against the Temporal Container
The container picture deserves explicit refusal, because it is not merely a loose way of speaking. It generates specific errors.
The picture holds that there is time, and that within time there are processes, some fast and some slow. Time is the neutral dimension along which everything is arrayed, and any two processes can therefore be compared by locating them on it. Coordination is what happens as things move along this dimension together.
The first thing wrong with this is that the neutral dimension is not neutral. What presents itself as time in this picture is clock time, and clock time is not a feature of the mesocosm encountered as such. It is a stabilization with a describable history. Regularities were extracted from dwelling, the rotation of a planet and its orbit around a star, and were then stabilized poietically: materially, in gnomons, water clocks, escapements, pendulums, quartz oscillators and atomic resonators, and symbolically, in counted units, calendars, notations and eventually a definition detached from the astronomy altogether. Both poietic mediations are involved throughout, and a clock is as much a material technology as a symbolic one. What arrives at L4 as clock time is the product of that whole apparatus, and its success is exactly what allows the apparatus to disappear and the product to present itself as the medium rather than as a stabilization.
The second thing wrong is that treating time as a container makes coordination look easy. If everything is arrayed along one dimension, then processes at different rates are simply at different points on it, and there is no particular problem about their relation. But the problem is the whole phenomenon. A wound heals over weeks while the person is expected at work on Monday. A forest regenerates over centuries while a planning permission runs for five years. A child develops over eighteen years while a school year runs for nine months and a lesson for fifty minutes. In none of these cases is there any difficulty about locating the processes on a common scale. The difficulty is entirely in the relation, and the common scale contributes nothing to resolving it.
The third thing wrong is that the container picture makes temporal misfit look like a scheduling problem. If everything is on one dimension, misfit means things are at the wrong points on it, and the remedy is to move them. A great deal of institutional response to temporal misfit takes exactly this form: extend the deadline, shorten the wait, add a session. Sometimes this works. It works when the processes involved can in fact be relocated. It fails, often severely, when the rate is intrinsic to the process, when a body cannot heal faster because a schedule requires it and a relationship cannot be repaired in the time an institution allocates to repair.
What replaces the container is not a rejection of measurement. Clocks work, and this article uses them freely. What is rejected is the assumption that the measure is prior to what it measures, and that comparison along it is therefore always available and always sufficient.
III. Bandwidth, Not Tempo
The five mediations are commonly and loosely characterized by speed. Embodiment is fast, being-with is medium, dwelling is slow, multimateriality is durable, multisymbolization is enduring. This is a serious mischaracterization and it should be abandoned.
Embodiment is not fast. Embodiment ranges from the sub-millisecond kinetics of a synaptic event through sensorimotor adjustment at tens of milliseconds, cardiac and respiratory cycles at seconds, the roughly ninety-minute architecture of a sleep cycle, digestive processes across hours, circadian organization across a day, menstrual and seasonal rhythms across weeks and months, wound healing across weeks to months, bone remodelling across years, developmental transition across decades, and senescence across a whole lifetime. That is a span of roughly twelve orders of magnitude within one mediation.
Being-with is not medium. It ranges from the fractions of a second within which mutual gaze and turn-taking are organized, through the seconds of conversational repair, the days of a quarrel, the years of a friendship, the decades of a marriage, and the multi-generational reach of a lineage in which a grandmother's practice reaches a grandchild she coordinates with directly.
Dwelling is not slow. Dwelling includes a lightning strike, a gust, a wave, a squall, a tide, a frost, a season, an ecological succession, an ice age, an orogeny, and the age of a star. Its fast end is faster than most of what embodiment does.
The correct characterization is that each mediation has a temporal bandwidth: a range of rates across which it coordinates processes. Bandwidth is a property of the mediation. Rate is a property of a particular process within it. The two must not be conflated, and the loose speed-characterization conflates them by taking whichever rate is most salient in a given example and treating it as the mediation's tempo.
Note that the bandwidths overlap heavily rather than occupying separate registers. The fastest coordination in being-with and the fastest in embodiment sit in the same range, since conversational turn-taking and sensorimotor adjustment are both organized in the hundreds of milliseconds and depend on each other. Nothing in the bandwidth account requires one mediation to be quicker at its quick end than another. What it requires is that no mediation be assigned a single rate.
This reformulation does real work. It explains why so many attempts to align institutional practice with a mediation fail: they align to a point within the bandwidth and then encounter everything else in it. A health system organized around the timescale of an acute episode encounters chronic conditions running at the timescale of a life. A workplace organized around the day encounters a grief that runs for years and a startle response that runs for a fifth of a second, and has procedures adapted to neither. The mediation was never at one rate to be aligned with.
It also blocks a tempting inference. If mediations had tempos, one could ask which mediation is temporally dominant in a given coordination and expect the answer to settle how that coordination unfolds. Since mediations have bandwidths, the question has no answer of that shape. What can be asked instead is which rates within which mediations are actually being related to which others, which is a harder question and the right one.
IV. A Note on Distribution Within a Bandwidth
Bandwidth alone would suggest that each mediation is uniformly spread across its range, and it is worth registering that this is unlikely to be true. Within a bandwidth, coordination probably clusters. There are likely to be regions where a great deal of a mediation's activity concentrates and long tails where relatively little does.
This is offered as an empirical hypothesis rather than as a formal component of the architecture, and the article does not rely on it. No numerical distribution is proposed here, because none has been established, and proposing one would give a heuristic the appearance of a result. What can be said is that the distribution is a describable feature of any actual mesocosm, that it will differ between mesocosms, and that where a coordination sits relative to the crowded regions of its mediation is likely to be a diagnostic fact about it.
The reason to expect the hypothesis to be productive is that institutions grow up around crowded regions. Where a great deal of a mediation's coordination concentrates, practices, vocabularies and institutional forms accumulate to handle it. A coordination in a sparse region of the same mediation is then likely to be poorly supported and weakly articulated, not because anything about it is defective but because nothing was built for it. Very slow embodied processes and very fast relational ones plausibly share this predicament, and if they do, the shared predicament is explained by their position within a distribution rather than by anything they have in common substantively.
Establishing any of this requires empirical work that has not been done, and the concept should be held loosely until it has been. What follows in the next three sections does not depend on it. Those sections establish something considerably more secure: that the bandwidths are grossly unequal in extent, and that the inequality is structural.
V. Embodiment: The Lifetime Bound
Embodiment has the narrowest bandwidth of the five, and it has a feature none of the others has: a hard upper bound that is not merely statistical.
The upper bound is the lifetime. No embodied process extends beyond the death of the body that carries it, because there is no body to carry it. This bound is categorical rather than gradual. Embodiment's slow end does not thin out toward the centuries. It stops.
The bound is also, by any comparison outside itself, extremely short. Human beings are in fact unusually long-lived for mammals of their body size, and unusually long-lived after the end of reproduction, which is one of the more distinctive facts about the species. The point here does not depend on humans being ordinary. It depends on the comparison being made against dwelling rather than against other organisms. Even the extreme cases barely change the picture: a sequoia at two thousand years, a bristlecone at five thousand, a clonal system whose continuity may run to tens of thousands, all remain trivially brief against dwelling's slow processes. The sequoia that overwhelms a human observer with its antiquity has occupied roughly one four-hundred-thousandth of the age of the ground it stands on. Nothing alive has a temporal extent that registers on a cosmic scale.
This asymmetry is structural rather than curious, and it determines a great deal.
It means that embodiment is permanently outranged by dwelling. Every living being coordinates with conditions whose characteristic durations exceed its own by many orders of magnitude, and it coordinates with them from inside a window too short to observe them changing. A mountain is not stable. It is being lifted and eroded continuously. It presents as stable because the observing body's entire bandwidth fits inside what is, for the mountain, a moment. Dwelling's apparent permanence is a relation between two bandwidths rather than a property of terrain.
It means that the three zoetic mediations are collectively bounded above by the lifetime, with one qualification taken up in the next section. Coordination that must persist longer than a body has no zoetic mechanism available to it.
And it means that temporal depths beyond the lifetime are unavailable to zoetic mediation as such. A living being with only embodiment, being-with and dwelling has no access to what happened before it existed and no access to what will happen after. It coordinates with deposits left by the past without any means of recognizing them as deposits or as past.
Everything that follows about the poietic mediations should be read against this bound. Poiesis is, first and last, the extension of temporal coordination beyond a bandwidth that is otherwise very short.
VI. Being-With: Chained Lifetimes
Being-with is bounded above by the lifetime in the same way embodiment is, since both parties to a relation must be alive for the relation to proceed. But it has a property that changes what that bound implies, and the property is worth isolating because it is easy to overlook.
Lifetimes overlap. A person coordinates directly with people born decades before them and decades after. Each of those people did the same. The result is that being-with supports a chain of direct coordinations extending far beyond any single lifetime, without any durable external deposit being required at any point.
The chain has a characteristic reach. Direct being-with typically spans something like a century and a half in practice, since a person may coordinate as a child with someone very old and again in old age with someone very young, and the two endpoints of that span never coordinate with each other at all. Beyond that, transmission continues through further overlaps, and a practice can pass through many of them and persist for a very long time.
The qualification that matters is what the transmission does and does not deposit. It deposits a great deal: embodied skill, habit, posture, rhythm, mnemonic technique, formulaic structure, ritual sequence, and in many traditions elaborate and deliberate devices for stabilizing what is transmitted. Oral transmission is not the absence of stabilization. What it lacks is a durable external poietic deposit standing outside the chain and available for comparison. There is no earlier version to hold against the current one, and a version held only in living memory is itself a link in the chain rather than an independent check on it.
The consequence is not that drift is invisible. Elders can and do notice that something has changed, and traditions frequently contain explicit resources for saying so. The consequence is that drift becomes progressively harder to reconstruct as chains lengthen, because each reconstruction is performed with materials that have themselves been transmitted. Over short chains the comparison is available and often sharp. Over long ones it degrades, and the degradation is a structural feature rather than a failure of care.
This gives being-with a temporal signature neither poietic mediation shares, and it explains why oral and written continuity produce such different results. It would be a mistake, however, to describe what survives orally as what remained useful. Transmission is sustained by many things other than usefulness: authority, coercion, ritual obligation, prestige, pleasure, habit and inertia all transmit material that no adaptive account would predict, and a great deal that is transmitted is transmitted because someone with standing insisted on it. The difference from written transmission is not that oral transmission selects for use. It is that written transmission fixes a form that can be compared, and oral transmission does not.
At the fast end, being-with contains coordination organized at gaps of around two hundred milliseconds in conversational turn-taking, which is shorter than the time required to plan the utterance being produced. The response was therefore already under construction while the other party was still speaking. Anticipation at this scale is not an occasional feature of interaction but a condition of its possibility, and it will matter for the account of felt misalignment below.
VII. Dwelling: Found Conditions and Made Ones
Dwelling has the widest bandwidth of the five by an enormous margin, running from the microsecond of a lightning discharge to the multi-billion-year scale of stellar and planetary process. Its defining character within the architecture is that it is found rather than made: coordination with conditions on the counterpart's own uneven and uncontrollable terms.
A distinction is required here that the general characterization does not by itself supply. Some dwelling temporalities were not produced by life and do not depend on it. Planetary rotation, orbital period, the solar cycle, tectonic process and stellar evolution would proceed unchanged if nothing were alive. Others are found by any particular organism but were produced by life collectively: a soil, a forest canopy, an oxygenated atmosphere, a reef, a river course altered by beavers, a grassland maintained by grazing. These arrive as given conditions for the organism encountering them, and coordinate on the counterpart's terms exactly as terrain does, while having temporalities that living process itself established.
The distinction matters because the two classes behave differently. Cosmically found temporalities are indifferent to life and cannot be altered by it. Life-produced dwelling conditions can be altered, sometimes rapidly, and their rates are consequently contingent in a way the astronomical regularities are not. An atmosphere's composition has a history and a changeable future. A day does not.
Two further features of dwelling's temporal profile deserve emphasis.
The first is that its regularity and irregularity are distributed very unevenly, and life exploits this. The astronomical regularities are among the most predictable phenomena any organism encounters, and enormous amounts of biological organization are anchored to them; circadian organization exists because a planetary rotation existed first. Weather sits at the other extreme, governed by fixed laws with initial conditions no organism can track. Dwelling therefore offers simultaneously the best available temporal anchor and one of the least predictable temporal environments, and organisms handle the two through entirely different strategies, entrainment for the regular and responsiveness for the irregular.
The second is that dwelling's slowest processes are effectively unobservable from within any lifetime. An organism cannot watch an ecological succession or an orogeny. It occupies one point in them. Whatever changes at those rates presents to a living being as unchanging ground, and the presentation is systematically misleading. Much of what a mesocosm treats as permanent background is moving at a rate no living bandwidth can register.
This produces a specific form of misfit, and it is where the two classes of dwelling condition converge. When a slow process becomes fast enough to enter a living bandwidth, whether through natural variation or through cumulative human effect, it arrives as a violation of something that had never been treated as a variable. The coordination fails not because a rate was misjudged but because no rate had been assigned. Ground does not have a rate; it is what other things happen against. A climate changing within a lifetime is not a faster version of a familiar problem. It is the conversion of ground into process, and a mesocosm organized around the found permanence of dwelling has no established position from which to encounter it.
VIII. Multimateriality as Rate Transformation
The two poietic mediations extend, transform and transport temporal coordination beyond the limits of individual embodied enactment. This is the article's central claim about them, and it is not an added perspective. It is what they do.
Take multimateriality first. Its defining character is that it deposits prior recursive coordination into durable material organization. Every term in that characterization is temporal. Deposit is a temporal operation: something that was occurring is made to persist after the occurrence stops. Durable is a temporal property: a rate of change slow enough that the thing can be encountered later as the same thing.
The operation performed is not uniformly deceleration, and it would be a mistake to characterize it that way. What multimateriality does is transform rate relations, and it does so in several directions.
Most commonly it slows. A container makes food decay at a much lower rate than it otherwise would. A wall makes a boundary persist that would otherwise be renegotiated at every encounter. An archive slows the loss of a record.
Frequently it accelerates. A tool makes a process fast that was slow. A vehicle compresses a journey. A mill does in an hour what hands would do in a week. The point of acceleration is the same as the point of deceleration: bringing a process into a rate relation where coordination can proceed.
It also regularizes. A machine converts a variable process into a periodic one. A dam converts an irregular flow into a controlled one. A thermostat converts a fluctuating interior climate into a steady one. Here what changes is not speed but variance, and the value lies in predictability rather than in rate.
And it buffers, delays and synchronizes. A reservoir holds water until it is wanted. A warehouse holds goods across the gap between production and demand. A railway timetable synchronizes movements that would otherwise arrive in no relation to one another. In each case the operation is on phase and availability rather than on rate as such.
What unifies these is not a direction but a target. Multimateriality operates on the temporal relations between processes, taking a relation that does not work and installing one that does.
What it cannot do is escape rate. Every deposit changes. A wall requires pointing, a road resurfacing, an archive climate control. The manufactured stability is always a transformed rate rather than an arrested one, which is why maintenance is constitutive of multimateriality rather than incidental to it. A deposit that is not maintained is a deposit on its way back into dwelling, and the effort required to sustain one is proportional to the gap between its rate and the rate of everything around it. A cathedral maintained for eight centuries has consumed a great deal more coordination than the coordination it originally deposited.
IX. Multisymbolization and Temporal Portability
Multisymbolization deposits a distinction rather than a material organization, and this changes what persistence means without removing it from the rate structure.
The temptation here is to say that a symbolic deposit has no rate of change at all: that the mark decays but the distinction it carries does not, and that symbolization therefore achieves what multimateriality cannot. This is too strong, and it would reinstate exactly the symbolic ontology the architecture elsewhere resists.
What can be copied without loss is a form. A string of marks can be reproduced exactly, and this is a real and consequential property, unavailable to material deposits, whose copies are new deposits with their own decay. But a form is not yet a distinction. A distinction requires enactment: someone must read the marks, and reading is a living process that depends on training, on a language, on institutional context, and on a whole apparatus of interpretive practice that is itself transmitted and itself changes. The form persists. What the form does when enacted does not persist in the same way, and frequently does not persist at all.
The evidence for this is ordinary and abundant. Words are coined and lose their meanings. Legal texts are reinterpreted until they license what their drafters would have refused. Classificatory contrasts shift under identical vocabulary. Notations become unreadable. Scripts are lost. A term of art in one century names something quite different in the next while every letter stays where it was. In each case formal persistence was complete and enacted continuity was not.
The correct account therefore separates three things rather than two. There is the material carrier, which decays at a material rate. There is the form, which can be replicated across carriers without loss and is in that respect uniquely portable. And there is the enacted distinction, which requires a living being with a bandwidth and an inheritance, and which drifts.
Temporal portability is the right name for what multisymbolization achieves. A distinction, once drawn, can be carried across absence, delay, distance and death, and can be reconstituted by someone who was not present at its making. This is an extraordinary capacity and it is the basis of everything poiesis makes possible at the symbolic level. It is not the same as timelessness.
The characteristic problems of symbolic life follow from the gap between the three. A category deposited three centuries ago arrives with its form intact and gives no indication of how far its enactment has drifted from what it once did. Material deposits announce their age; weathering is legible. Formal persistence is silent about elapsed time, which is why a symbolic deposit can present as contemporary while the coordination it once tracked has moved entirely. The problem is not that symbols escape time. It is that the most portable component of a symbolization carries no temporal signature, while the component that does drift carries no external record against which the drift could be measured.
X. Mathematics as Formal Invariance
Mathematical symbolization is the extreme case of formal portability, and examining why yields a derivation rather than an observation, provided the claim is stated with the restrictions the previous section requires.
A mathematical structure functions, within the architecture, as a purely nonrecursive symbolic counterpart. Nothing about how it is engaged alters how it proceeds. This is what distinguishes it from other symbolic systems and is the reason its applicability to any domain must be established case by case against a cosmos that owes it nothing.
Now consider what temporality means here. Temporality is relational structure among unfolding processes. A counterpart that never alters course in response to being engaged presents no unfolding to coordinate with. It has no trajectory, no phase, no rate. Apparent atemporality is therefore not a separate metaphysical status that some entities enjoy. It is what maximal nonrecursivity looks like from inside a temporal process.
That 2+2=4 does not age is accordingly not evidence that arithmetic inhabits a timeless realm. It is what invariance within a stabilized formal system looks like when encountered by living beings whose every other counterpart alters. A word can be coined and lose its meaning because words are embedded in coordination that continues to move and continues to reach back and alter what the word does. The relation the arithmetic expresses is not embedded in that way, and is not available for renegotiation, which is why it can be relied on and why it can be applied to domains that owe it nothing.
The restriction is essential and should be stated flatly. What is invariant is the formal relation within a stabilized system. Nothing else about mathematics is invariant. Notations are invented, adopted and abandoned. Axiomatizations are replaced. What counts as a proof has changed repeatedly and is changing now. Concepts are extended until they mean something their originators would not recognize, and number itself is a case in point. Pedagogical transmission, disciplinary practice, standards of rigour and the institutional life of mathematics all have thoroughly ordinary temporal profiles. The mathematical counterpart and the multisymbolic practices through which it becomes available are different things, and only the first is invariant.
The philosophical afterlife of the confusion is worth naming, because it is a clean instance of a general hazard. Symbolic practice hides its own operations. The enormous infrastructure supporting mathematical work, the training, the notation with its documented history, the community of verification, the paper and the boards and the bodies, disappears into L1 and does not surface. What remains salient is the invariant relation. The inference from an unmoving relation to a realm of unmoving objects is then very easy to make, and mathematical Platonism is what that inference looks like when it is made carefully and defended at length.
The alternative offered here is neither Platonism nor a reduction of mathematics to convention. The invariance is real and is not conventional; nothing about agreement makes it hold. What is mistaken is only the step from the invariance of the relation to a claim about where the relation resides, and the step is made easy by the disappearance of everything that had to happen for anyone to encounter the relation at all.
A practical consequence follows for any domain that borrows mathematical form. Importing a formally invariant structure into a coordination that is not invariant imports the stillness with the structure. The formalism will not register that what it describes has moved, because registering movement is not something it can do.
XI. What Clock Time Cannot Exhaust
It should now be possible to state a claim carefully that would be false if stated absolutely. Clock time cannot exhaust the temporal structure of the mediations, and it cannot serve as a neutral account of temporal coordination.
It can compare them, and this article has done so throughout. Milliseconds, seconds, days, years, centuries and billions of years have been used to establish bandwidths and orders-of-magnitude differences, and those comparisons are informative and correct. Nothing here recommends abandoning measurement. What is at issue is what measurement delivers and what it leaves out.
Clock time measures duration and rate, and it measures them extremely well. Where duration and rate are what matters, it is the right instrument and there is no substitute. This is why physiology is so measurable, why the manufacture and use of made things have determinate durations, and why dwelling's regularities are describable with such precision.
What clock time does not measure is everything else in the relational structure. It is silent about order, since two processes of equal duration can stand in either order. It is silent about phase, since identical durations are compatible with any phase relation whatever. It is silent about nesting, anticipation, recurrence and interruption. And phase, not duration, is usually what determines whether a coordination works. A meeting's hour and a person's alertness cycle have a phase relation; the meeting's length tells you nothing about it.
Two further limitations follow from where clock time came from. It originated as an extraction from dwelling, stabilized through material and symbolic technology, and it consequently fits dwelling best and fits the periodic portions of embodiment well. It fits being-with poorly: the duration of a conversation in minutes is close to uninformative about what the conversation accomplished, and two years of one relationship and two years of another are not comparable quantities in any respect that matters. It fits the nonperiodic portions of embodiment poorly, which is why a healing time is always a distribution rather than a value. And it says nothing about symbolic portability, since the portable component of a symbolization has no duration and the enactment's duration is a fact about the enacting body.
There is also a political consequence, and it is not incidental. Once clock time is installed as the universal measure, coordination whose temporal structure is not primarily duration gets recoded into duration at the point of institutional recording. Care work measured in minutes per visit, education in contact hours, therapeutic work in sessions, and grief in days of leave are all instances. In each case a coordination organized around phase, responsiveness and recurrence is recorded as an interval, and everything about it that is not interval is lost when it is written down. The measure is not wrong about what it measures. It is that what it measures is not what was at stake.
XII. What Can Be Said Instead
If duration is one element of temporal structure rather than the whole, the other elements need to be nameable. They are, and they are describable without a shared unit. This relational catalogue is what a temporal account of a mesocosm should largely consist of.
Order can be established. Whether one process begins before another, whether a response follows or precedes what it responds to, whether a deposit predates the coordination encountering it: these are determinate and often decisive, and none requires a common unit.
Phase can be established. Whether two processes meet at workable points in their respective cycles is the central temporal question about most coordination and is independent of their durations. A tide and a fishing practice have a phase relation. So do a crop cycle and a labour supply, an infant's sleep and an adult's, an institution's reporting period and the process it reports on.
Nesting can be established. Whether one process is contained within another, and how many cycles of one occur within one cycle of another, are ratio facts. Ratios survive the absence of a shared unit because they are relations between processes rather than between a process and a metric.
Rate relation can be established. Whether one process is faster than another, and by roughly how many orders of magnitude, is usually determinable and usually sufficient. That embodiment is outranged by dwelling by many orders of magnitude does not require either to be measured precisely.
Persistence relation can be established. Whether a deposit outlasts its makers, whether an institution outlasts the coordination that produced it, whether a practice survives the generation that learned it, are yes-or-no questions with real answers.
Anticipation can be established. Whether a coordination is organized around an expected continuation, and what that expectation consists in, is describable and is frequently the whole of what is going on, as conversational turn-taking demonstrates.
Recurrence and interruption can be established. Whether something happens again, whether it happens when expected, whether it stops, whether it resumes, are among the most consequential temporal facts about any coordination and none is an interval.
Where clock measurement is available and appropriate it should be used, and these relations are often easiest to establish with a clock in hand. What should be resisted is the assumption that a coordination has not been temporally described until a number of units is attached to it. The relational description is frequently the complete one, and attaching units adds precision about a dimension that was not in question while contributing nothing about the ones that were.
XIII. Temporality and the Independent Properties
An objection arises here which needs answering directly, because if it succeeds the article has been restating something already in place.
Stability already names the rate and extent of a counterpart's own variation, and it is already held independent of recursive type, predictability, controllability and durability. A bone is stable and a heartbeat variable, both selfrecursive; a mountain is stable and weather variable, both nonrecursive. If stability already names rate of change, what does differential temporality add?
The answer is that stability is a property of a counterpart and temporality is a property of a coordination, and the difference is not a matter of emphasis.
Stability is monadic. It can be assessed by observing one counterpart. Bone changes slowly; that is a fact about bone available without reference to anything bone coordinates with.
Everything in the relational catalogue is not monadic. Phase is a relation between two processes. Nesting is a relation. Anticipation is a relation between a trajectory and its expected continuation. Interruption is a relation between an unfolding and its cessation. Bandwidth is a property of a mediation rather than of any counterpart, describing the spread of rates across which that mediation coordinates. None of these is recoverable from the stability values of the counterparts involved. Two highly stable counterparts can stand in a disastrous phase relation, and two rapidly variable ones can be exquisitely coordinated.
The second difference is that stability is a rate and the relational catalogue contains structure that is not rate at all. Recurrence, order, anticipation and interruption are not faster or slower versions of anything. A process that recurs and a process that does not may have identical rates.
The third difference is the one drawn in section I. Stability is one property among several listed as independent. Differential temporality is not a further item for that list, because it is not a property counterparts have. It is the condition that makes continuity an achievement rather than a given. Adding it to the list would be a category error of the same kind as adding continuity itself.
The properties therefore stand as they are, and their relation to this article is straightforward. Stability contributes to a counterpart's temporal profile without exhausting it. Durability names how long a formed state persists, which is one temporal fact among several that matter about a deposit and says nothing about the deposit's phase relation to the coordination encountering it, which is frequently what determines whether the deposit helps or harms.
XIV. The Levels Are Not Durations
The recursivity levels have sometimes been read as though they carried characteristic timescales, with L1 as the ongoing, L2 as the momentary, L3 as the occasional, and L4 as the enduring. This reading should be rejected, and rejecting it yields the article's central result about the architecture.
L1 can last a hundred milliseconds or seventy years. L2 can be a flash of ten milliseconds when a foot meets a step that is not there, or an unresolved sense of wrongness carried for a decade. L3 can be a two-second utterance or a research programme. L4 can be a classification stabilized in a committee meeting and abandoned within the year, or one that organizes a civilization for a millennium. L5 can be a moment of repositioning or an entire intellectual tradition.
The levels are an architecture of recursive availability rather than a clock. They describe whether coordination is proceeding transparently, becoming felt, becoming articulated, becoming stabilized, or becoming available for repositioning. None of these is a duration and none has one.
It is worth adding that enactment and deposit are orthogonal to level, and should not be used to supply the bound that duration fails to supply. Every level can be enacted, and articulations at several levels can be deposited. An L3 articulation can be written down. An L4 classification is enacted every time a case is assigned to it, and much of what institutions do is precisely this enactment. L5 repositioning is characteristically enacted, and although an L5 insight can be deposited, the deposit is not thereby an ongoing act of L5. Whether something is being enacted now or was deposited earlier is a real and important distinction, and it cuts across the levels rather than along them.
The consequence that matters most concerns L1 and L2, and it resolves a difficulty that has looked intractable. No duration threshold has ever successfully separated them, and various attempts to find one have failed. They failed because they were looking for the wrong kind of thing. The distinction is not between a longer and a shorter process. It is between coordination proceeding without becoming separately salient and a difference within its own unfolding becoming felt.
Stated more fully: at L1, processes at heterogeneous rates stand in relations that hold well enough that the differences among them require no registration. At L2, something in that relation changes. One process outruns another, arrives late, persists past its expected end, stops, fails to recur, recurs when it should not, or diverges from what preceded it. The difference within the temporal relation is what becomes felt.
This gives the porousness of the L1/L2 boundary a precise account without converting it into a hard division, and it explains why the boundary is porous. The relations among rates vary continuously, and there is no threshold at which variation becomes difference. It also explains why duration was never going to work as a criterion: an L2 is not a length of anything.
Whether every L2 has this structure, including cases that appear purely spatial or purely qualitative, is a further question requiring its own treatment, and the article does not settle it. Sudden pain, colour mismatch, disgust, unfamiliarity and social awkwardness would all need examination, and the result would be substantial either way. What is established here is the weaker and still consequential claim: that the levels are not durations, that enactment and deposit do not supply an alternative bound along the levels, and that the temporal relation between L1 and L2 is one of transparency and felt difference rather than of length.
One distinction should be recorded before leaving this, because it bears on how sustained disturbance is described. An L2 episode and a condition that keeps generating L2 episodes are not the same thing. Someone whose pain has persisted for years is not in a single continuous state of felt misalignment: there is sleep, there is attentional withdrawal, there is habituation, and salience characteristically returns at variations rather than at peaks, which is what a difference-within-unfolding account predicts. What persists is a condition that keeps regenerating the disturbance, together with a body being depleted by the regeneration. Whether sustained recursive salience has a specifiable ceiling, and of what kind, is a substantial question that would require its own treatment and is not settled here.
XV. Polyrhythm, Temporal Fit, and Resynchronization
If continuity is the achievement of workable relations among heterogeneous processes, the character of that achievement can be specified.
It is not the imposition of a common tempo. Living coordination does not work by bringing everything to one rate, and the attempt to do so is a recognizable pathology rather than an ideal. Coordination is polyrhythmic: it holds processes at different speeds, durations, periodicities and phases in relations that work, without any of them being brought into line with the others.
Synchronization, in this sense, does not mean simultaneity. An orchestra is synchronized without every instrument playing the same note at the same moment. Rest can fit activity, waiting can fit anticipation, one person speaking can fit another listening, and a fallow season can fit a growing one. What synchronization names is appropriate phase relation, and phase relations include alternation, delay, anticipation, recurrence, pause and interruption as fully as they include coincidence.
Temporal fit can accordingly be understood as a dimension of mesocosmic fit rather than as a further category alongside it. Mesocosmic fit already names the degree to which an articulation or arrangement tracks the coordination it addresses across the five mediations. Whether heterogeneous processes meet at workable moments and rates is part of what that tracking consists in, and separating it out as an independent kind of fit would obscure how much mesocosmic misfit has a temporal component.
That component has a recognizable grammar. Too early. Too late. Too fast. Too slow. Too long. Not long enough. Interrupted. Never arriving. Repeated when it should not be. Absent when it should recur. This vocabulary describes misfit across domains that otherwise have nothing in common, from sleep to bereavement to institutional process, and its cross-domain reach is evidence that the temporal structure is doing work rather than being read into the cases.
The final point is the most important for any normative use of this account. Perfect temporal fit is not available. Rates diverge continuously, and every coordination is perpetually being desynchronized by processes it does not control. What distinguishes a viable coordination is therefore not the maintenance of fit but the capacity to recover it: to absorb disturbance, change pace, wait, accelerate, interrupt, resume, and establish new workable relations after the old ones have failed.
This connects temporality to recursive fluidity. Fluidity is the capacity to move between levels in ways that preserve and repair coordination, with the aim of returning to L1 with better coordination rather than remaining in indefinite reflection. Resynchronization is what that return looks like when the disturbance was temporal, which is a large proportion of the time.
Where resynchronization fails, the reasons vary and no single diagnosis covers them. Sometimes a rate has been fixed by something that will not move, and asking what will not move is then a powerful question. But failure also follows from irreversible loss, from the death of a participant, from resource depletion, from stochastic instability in processes that were never regular, and from trajectories that have become incompatible without any of them being fixed. Asking what will not move is one diagnostic among several, and treating it as the diagnostic would misdescribe a great many failures.
XVI. Consequences
Several things in the architecture look different once differential temporality is taken as constitutive rather than additional.
Value looks different. Value is what sustains, repairs, intensifies or improves coordination. If coordination is polyrhythmic achievement, then a large part of what sustains and repairs it is temporal: the maintenance of workable phase relations, the absorption of desynchronization, and the recovery of fit after disturbance. This is not a redefinition of value but a specification of what a great deal of valuable activity does, and it identifies a class of work, much of it in care, maintenance and repair, whose product is precisely the continued holding-together of processes that would otherwise diverge.
The muting of coordination looks different, and requires a qualification that is easy to miss. Coordination that fully meets a stake tends to disappear from view, and temporal coordination often disappears completely, since when relations hold there is frequently nothing to notice. But it would be wrong to say that successful temporal fit is the absence of any temporal phenomenon. Dance, music, chanting, skilled sport, sexual coordination and deliberate breathing all involve intensely salient temporal coordination that is going extremely well. What these show is that salience and misfit are separable: L1 can be rhythmically rich without being temporally problematic. The correct claim is narrower and still substantial. Successful temporal fit need not become salient as a problem, and where it does not, the work of maintaining it becomes invisible even by the standards of invisible work, which is why such work is so consistently discovered only through its failure.
Power looks different. Control over the temporalities of others is a distinct form of power, separable from control over their resources or their categories. The capacity to set the rate at which someone else must operate, to impose waiting, to demand response within an interval, to schedule, and to defer indefinitely is exercised continuously and is rarely named as power at all, because it presents as administration.
The asymmetry underlying that power can now be stated precisely. Recursive disturbance must be enacted by a body: it is bodies that feel misalignment, and enactment draws on a body's resources. Poietic arrangements need not be enacted by any particular body to persist. A form, a rule, a schedule or a threshold is deposited, and it can go on imposing the same demand across successive bodies indefinitely. This does not make institutions costless. They require personnel, energy, records, buildings, software and enforcement, and their maintenance is as real as any other deposit's. What it makes them is differently costed. The cost is displaced and distributed rather than borne where the demand lands, and the deposit itself does not tire in the way a body tires. Any account of what institutions do to people that lacks this asymmetry will keep describing the resulting exhaustion as individual failure of resilience.
Poiesis looks different. Both poietic mediations extend temporal coordination beyond what embodied enactment allows, and they do it by different means: multimateriality through durable material organization whose rate relations can be engineered, multisymbolization through replicable form whose portability outruns the drift of what it enacts. And both operate under an asymmetry nothing reverses. The past can be deposited. The future can only be projected, promised, scheduled and modelled, and every such projection remains open to revision by what actually arrives.
That asymmetry is where an account of temporality in this architecture properly ends, because it is the condition under which every other temporal operation is performed. Living beings coordinate forward into a future they cannot deposit, using ground deposited by a past they did not witness, from inside a bandwidth too short to observe either. Continuity, under those conditions, is not a background fact about the passage of time. It is the most ordinary and least noticed of achievements.