WE INHERIT THE Material historicity, consciousness |
|---|
TRACES · NORMATIVITY · VALENCE · REFLEXIVITY · CULTURE
Daloze Didier
ORI-C: ori-c.be
July 2026
STATUS OF THE TEXT A theoretical proposal and an interdisciplinary research programme. The framework connects material historicity, living organisation, consciousness and freedom without supposing an elementary consciousness in all matter and without introducing any substance foreign to the physical world. |
|---|
« Matter is historical. The living exploits that historicity.
Consciousness undergoes it. Reflexivity renders it partly revisable. »
Abstract (French version)
This article proposes a historical and organisational naturalism of consciousness. Its starting point is simple: matter is never without a past. Its interactions can leave persistent modifications that influence later transformations. The living does not create this historicity. It organises it into mechanisms of regulation, functional memory and viability. From this base, the article distinguishes three interdependent branches: a functional branch, running from trace to metacognition; a phenomenal branch, which explicitly poses the lock of the passage from functional processing to lived experience; and a collective branch, in which traces are externalised into language, techniques and institutions.
This architecture avoids two symmetrical reductions. It does not dissolve consciousness into all matter, but neither does it make it spring from a matter presumed neutral, instantaneous and devoid of relevant properties. It defines freedom as a causality become recursive: a historical system can partly represent certain causes that organise it and make that representation one of the causes of its future evolution. Collective freedom then depends not only on the memory available, but on the distribution of the power to inscribe, name, interpret and revise common traces. The framework remains open on the hard problem and issues in empirical predictions bearing on trace, valence, metacognition, normative revision and institutional revisability.
Keywords: material historicity, trace, normativity, valence, phenomenal consciousness, metacognition, freedom, recursive causality, culture, politics of traces.
Abstract
This article develops a historical and organizational naturalism of consciousness. Its starting point is that matter is never without a past: interactions can leave persistent modifications that alter subsequent transformations. Living systems do not create this historicity but organize it through regulation, functional memory, and viability. On this basis, the article proposes a ramified architecture with three strongly interacting yet non-deducible branches. The functional branch follows the exploitation of traces from biological normativity and functional valence to learning, preference, counterfactual anticipation, metacognition, and causal revision. The phenomenal branch isolates the unresolved transition from functional prioritization to felt valence, situated experience, and continuity of a first-person point of view. The collective branch examines how language, technologies, institutions, and cultural practices externalize traces and distribute the power to inscribe, name, interpret, contest, and revise them. The framework avoids both panpsychism, which risks dispersing consciousness throughout matter, and dualism, which makes experience foreign to physical organization. It also reframes freedom as recursive causality rather than exemption from causation: a historical system becomes relatively free when it can represent some of the causes that organize it and make this representation itself one of the causes of its future transformation. A multi-level criterion of normative coherence constrains this revision by taking into account viability, relational effects, temporal reach, reversibility, contestability, and the preservation of open futures. The proposal explicitly preserves the phenomenal gap and formulates testable predictions concerning the causal efficacy of traces, dissociations between functional and phenomenal valence, cross-task metacognitive generalization, durable reflective revision, and institutional revisability. It therefore offers a common research language for organizational biology, neuroscience, philosophy of mind, and the social sciences without reducing one domain to another.
Keywords: material historicity, trace, biological normativity, valence, phenomenal consciousness, metacognition, recursive causality, reflective freedom, culture, politics of traces.
Introduction: a consciousness born of a matter that is never blank
Consciousness is often enclosed within an unsatisfying alternative. Either a mental dimension would already be present, to an infinitesimal degree, in all matter; or consciousness would arise late within the living out of a physical world entirely foreign to it. The first position risks extending the word consciousness until it loses its discriminating power. The second leaves open a rupture that is hard to think: how could a lived experience appear within matter described as devoid of all temporal depth, of all value and of all interiority?
A third route consists in displacing the question. Instead of first asking what matter is in itself, one may examine what it does, what it preserves and the way its transformations modify its future possibilities. This displacement does not by itself resolve the enigma of feeling. It does, however, reveal a neglected continuity: consciousness does not develop within a matter without a past, but within organisations constituted by traces, constraints and inherited regimes of transformation.
GUIDING THESIS Consciousness must not be sought as a substance hidden within matter. It can be studied as a regime in which a historical material organisation acquires increasing access to certain traces and constraints that constitute it, to the point of being able partly to revise their influence. |
|---|
1. Inheriting chaos: from the strong formula to the precise mechanism
1.1 A constitutive continuity, not a biological transmission
In one sense, every system whose present state depends on the trajectory it has followed carries a trace of its past. Systems far from thermodynamic equilibrium often render this historicity particularly visible through metastable configurations, persistent correlations or durable transformations.
The brain pushes this capacity to a remarkable degree. It does not merely preserve traces. It selects them, reactivates them, combines them, projects them into possible futures and, in some cases, integrates them into a conscious experience.
Memory therefore does not suddenly arise with the brain. The brain does not invent the trace. It deploys an architecture able to select, reactivate, integrate and revise it. It thus transforms the historicity of material organisations into functional, represented and sometimes conscious memory.
It is in this sense that we may say we « inherit the chaos of the Universe ». This is not a matter of information called chaos having been transmitted from generation to generation, but of a constitutive continuity. Organisms are made of a matter that remains sensitive to fluctuations, constraints, non-linearities, regime transitions and the history of its interactions. The living never leaves this dynamic ground. It stabilises some of its processes, damps others, and can exploit instabilities to explore new configurations.
The term chaos must here be distinguished from mere disorder and from randomness. A chaotic system may be deterministic while being strongly sensitive to its initial conditions. A bifurcation designates a qualitative change of regime when certain constraints vary. A fluctuation is a variation. A trace is the persistent effect of certain variations. A memory presupposes that this effect can be reactivated and used. The article therefore uses the word chaos as a condensed formula for a material world never perfectly fixed, in which organisation results from provisional stabilities and history-dependent transformations.
1.2 The living does not suppress instability: it channels it
Life does not constitute an island removed from physical dynamics. It maintains its organisation thanks to continuous flows and regulatory loops. It must preserve enough stability to persist while remaining plastic enough to respond to perturbations. This tension is central: without conservation, no continuity is possible. Without variability, no adaptation is possible.
Organisational approaches in biology describe living autonomy in terms of networks of constraints that mutually produce and maintain one another. A biological function is then not merely a useful effect observed from outside. It takes part in the conservation of the organisation on which it depends [1, 2, 22]. This framework makes it possible to speak of biological normativity without supposing a conscious intention: certain transformations are favourable or unfavourable relative to the system's domain of viability.
Work on brain dynamics further suggests that conscious waking states display regimes further from equilibrium and a richer dynamic exploration than several states of unconsciousness. Other results relate the loss of consciousness to a departure from certain critical signatures [10, 11, 25, 26]. These results do not establish that consciousness would be produced by chaos or criticality. They show more modestly that the richness of experience is compatible with a dynamic organisation reducible neither to rigidity nor to random noise.
2. Operational definitions
The soundness of the framework depends on terms precise enough to be related to observations. The following definitions do not claim to exhaust the concepts. They fix the minimal conditions of their use in this article.
| Concept | Operational definition |
|---|---|
| Trace | Relatively persistent modification of the state or organisation of a system, produced by a past interaction and able to modify the probability or the form of future transformations. |
| Reactivation | Return to causal efficacy of a trace within a new situation. |
| Biological normativity | Regulatory discrimination between states that maintain or degrade a domain of viability defined for a given organisation. |
| Functional valence | Global weighting that modifies the priorities of attention, learning and action, without prejudging any feeling. |
| Phenomenal valence | The actually pleasant, unpleasant or affectively neutral character of a lived experience. |
| Causal revision | Transformation by which the present use of a trace or a criterion modifies its weight, its effects or its conditions of future reactivation. |
| Multi-level coherence | Evaluation of a transformation according to its consequences across several interdependent systems, relations and temporal horizons. |
3. A ramified architecture rather than an automatic ladder
The first version of the model could suggest a regular ascent: matter, the living, sensitivity, cognition, consciousness, culture. That representation is too simple. A sophisticated regulation does not prove a lived experience. An experience may exist without elaborate metacognition. An institution may preserve traces without understanding them. Three branches must therefore be distinguished, which cross without automatically following from one another.
Figure 1: ramified architecture. The functional branch organises access to and exploitation of traces. The phenomenal branch poses the question of lived experience. The collective branch externalises traces and distributes the power to revise them.
READING THE FIGURE: MATERIAL HISTORICITY Functional branch: trace → regulation / viability → functional valence → learning → preference → counterfactual anticipation → metacognition / causal revision Phenomenal branch: bodily integration → phenomenal valence → situated experience → continuity of point of view Collective branch: externalisation of traces → confrontation of interpretations → institutional revision → politics of traces Cross-interactions: the branches influence one another without automatically following from one another. |
|---|
The functional branch follows the way a trace becomes regulation, learning, preference, anticipation and then causal revision. The phenomenal branch concerns bodily integration, felt valence, situated experience and the continuity of a first-person point of view. The collective branch describes the externalisation of traces into language, techniques and institutions, together with the conflicts over their inscription and their interpretation.
The crossings constitute precisely the open questions. How does a functional priority become a pleasant or painful experience? To what extent does metacognition require phenomenal consciousness? How do cultural tools transform individual access to one's own causes? The model does not efface these difficulties: it localises them.
4. Functional branch: from trace to causal revision
4.1 Biological normativity: regulating relative to a viability
A rock may preserve a deformation without using that deformation to maintain its organisation. The living introduces a different regime: certain traces and signals modify the system's responses relative to the conditions of its continuity. Biological normativity is not a conceptual judgement. It is a performative discrimination inscribed in metabolic networks, damage thresholds and feedback loops.
The chemotaxis of Escherichia coli illustrates this passage. The bacterium does not merely respond to an instantaneous concentration. Its sensory network compares recent variations, and adaptation by receptor methylation takes part in a short-term biochemical memory. Classic experiments show that wild-type cells make temporal comparisons over several seconds, while certain mutants deficient in adaptation lose that capacity [3, 4]. The trace does not thereby become a conscious representation. It acquires a function in orienting behaviour.
This normativity is always relative to a system, a scale and a temporal horizon. Apoptosis may be unfavourable to the cell but favourable to the organism. An inflammatory response may protect in the short term and become destructive if prolonged. To speak of « good » or « bad » therefore requires specifying for which organisation, over what duration and at the price of what consequences for the interdependent levels.
4.2 Functional valence: making certain variations a priority
Biological regulation may remain distributed among many local loops. With nervous systems, certain bodily and environmental information can be coordinated more globally. Functional valence designates this weighting: a state acquires a priority that orients attention, approach, avoidance, learning or inhibition.
This priority must not be equated with felt pleasure or pain. A system may treat a signal as urgent without our knowing whether it feels anything. One must also avoid directly identifying pleasure with viability. An immediate reward may reinforce a harmful behaviour. A pain may protect. An unpleasant effort may improve future viability. Functional valence is a situated estimate of the significance of a state for the system, not an infallible measure of biological good.
4.3 The bridge between valence, learning and preference
Learning mechanisms transform gaps between expected and obtained outcomes into durable modifications of preference. The phasic activity of dopaminergic populations has been related to reward prediction errors: an unexpected reward increases the signal, an expected reward shifts it toward the cue that announces it, and its omission produces a decrease at the expected moment [5]. This mechanism provides a precise example of how a present interaction reorganises future responses.
It would nevertheless be incorrect to reduce pleasure to dopamine. Research on reward distinguishes learning, motivational salience or « wanting », and hedonic impact, the pleasure felt. Dopaminergic systems contribute strongly to motivation and learning, while hedonic experience depends on partly distinct architectures [6]. This dissociation is important: it shows that a priority of action, a motivation and a feeling are not three names for one and the same mechanism.
Cognitive preference appears when learned values orient the selection of actions. Decisions may rest on consolidated associations without explicit simulation, or on internal models representing the relations between states, actions and consequences. Work distinguishing model-free from model-based influences shows that human choices generally combine both regimes [7]. The first form exploits accumulated values. The second renders absent trajectories causally present in the decision.
4.4 Counterfactual anticipation: when several futures act in the present
A physical bifurcation means that several regimes are possible according to conditions. Counterfactual anticipation adds a new operation: the system represents several continuations that are not yet realised and compares them before acting. In the rat, sequences of hippocampal activity can represent, before a movement, trajectories directed toward remembered goals [8]. These sequences do not prove a conscious imagination comparable to the human one, but they show how absent paths can take part in present selection.
The extension of causal reach is decisive here. The present state is no longer modulated only by the consequences of the past. It is also modulated by anticipated future consequences. The organism can bear an immediate cost for a deferred benefit, inhibit an attraction because of a foreseen danger, or explore an option because it reduces an uncertainty. Historical matter thus becomes an organisation able to render part of its possible future functionally active.
4.5 Metacognition and causal revision
Metacognition designates the capacity to evaluate certain cognitive operations: knowing that one is uncertain, estimating the reliability of a memory, or seeking missing information. In humans, metacognitive precision can be partly dissociated from first-order performance, and certain prefrontal lesions selectively impair the evaluation of one's own perceptual decisions [16]. Experiments in macaques likewise indicate a capacity to avoid memory tests when the available information is insufficient [17, 23]. In rats, several paradigms have been proposed to study uncertainty and information seeking, but their metacognitive interpretation remains debated [24], since first-order associative strategies must be excluded.
Reflexive normativity requires more than an uncertainty signal. It presupposes that a system can represent certain criteria organising its evaluations, compare their consequences, modify their hierarchy or their domain of application, and then transfer that modification to new situations. This is not an entirely self-constituted normativity: the subject chooses neither its initial body, nor the whole of its history, nor the cultural categories within which it learns to understand itself. Reflexivity is a partial capacity of access and revision.
5. Phenomenal branch: the lock of lived experience
5.1 Functional valence and phenomenal valence are not equivalent
The functional branch explains how a state can receive a priority, modify learning and orient action. It does not demonstrate why that state would be felt as painful, pleasant, disquieting or soothing. The passage from functional valence to phenomenal valence constitutes an autonomous lock, and not a detail that more complexity would mechanically suffice to fill.
Pain asymbolia provides an instructive clinical dissociation. Patients can identify noxious stimulation while displaying profoundly atypical affective and defensive responses [9]. The exact interpretation of these cases remains debated, but they show at a minimum that nociceptive detection, the identification of a stimulus and its affective reach are not one single operation. An architecture of consciousness must therefore distinguish processing, functional value and lived quality.
PHENOMENAL LOCK The framework explains how a system can treat a state as favourable or unfavourable, learn from it and orient its behaviour. It does not yet deduce why that functional polarisation is accompanied, in certain systems, by a « what it is like » to be that system. |
|---|
5.2 A situated, bodily and continuous experience
Phenomenal consciousness does not seem to be a mere addition of signals. An experience appears situated in a body, organised around a present and structured according to a point of view. The integration of external, interoceptive, mnemonic and affective signals contributes to this lived scene. This unity does not imply an immaterial centre. It may result from a dynamic and temporary coordination of distributed processes.
Several contemporary theories attempt to specify the neural conditions of conscious access or of an integrated experience: global neuronal workspace, integrated information, recurrent processing, higher-order theories and predictive approaches [12–14]. A direct experimental confrontation published in 2025 tested several predictions of the global workspace and of integrated information theory. The results supported certain elements of both frameworks while putting several central predictions in difficulty. No theory provided a definitive resolution [15].
This uncertainty imposes a conceptual discipline. The phenomenal branch must not be automatically deduced from the functional branch. A regulation may exist without proof of lived experience. A sentient experience may not require reflection on its own criteria. A behavioural metacognition can sometimes be explained without supposing a rich reflexive consciousness. The ramified architecture precisely makes it possible to represent these dissociations.
6. Reflexive normativity: revising the criteria of revision
6.1 From preference to the examination of the preferable
Cognitive preference selects between options according to criteria already active: security, reward, habit, belonging, coherence with a goal. Reflexive normativity begins when some of those criteria themselves become objects of examination. The system can ask whether what gives it immediate satisfaction is compatible with its long-term continuity, whether an inherited fear still corresponds to the present danger, or whether an internalised cultural norm remains defensible.
This operation does not create a will without a cause. A criterion is always examined on the basis of other traces, values or anticipations. The regress of meta-criteria therefore cannot be halted by an absolute foundation without reintroducing a dogma. It can, on the other hand, be transformed into a constrained recursion: each revision is submitted to a wider field of consequences and remains itself revisable.
6.2 Multi-level coherence as a pragmatic safeguard
A normative transformation gains coherence when it improves a local capacity without transferring a disproportionate cost onto other levels, other beings or other temporalities. This principle does not supply a calculable universal morality, but a set of operational questions. Does the revision preserve the minimal biological conditions? Does it integrate the consequences for the persons and environments concerned? Does it remain contestable and, if possible, reversible? Does it maintain a plurality of futures, or does it irreversibly lock the system into a fragile trajectory?
This perspective distinguishes three notions often confused. Plasticity is the capacity to change. Autonomy is the capacity to intervene in some of the conditions of one's change. Normative coherence is the capacity to evaluate that change across several relations and temporal horizons. A highly plastic system may be easily manipulated. An autonomous system may pursue a destructive goal. A viable freedom requires that revision integrate the dependencies that make action possible.
7. Freedom as recursive causality
7.1 No longer governed solely by the most immediate cause
Freedom does not require a supernatural rupture of physical laws. It may be defined as the gradual capacity of a historical system to make some of its determinations bear upon other determinations. An anticipation can inhibit an impulse. A reconstructed value can modify the effect of an old norm. A new experience can recontextualise a fear. A decision can transform the environment so as to make a future action easier or harder.
Causality does not disappear: it becomes recursive. Inherited causes produce a partial representation of their own influence. That representation constitutes a new material and symbolic configuration. It then modifies the conditions under which the initial causes are expressed. Consciousness does not stand above the brain like an external judge. The judge, the evidence, the rules and the conflict belong to the same system.
Freedom is not the absence of determination. It is the possibility of no longer being exclusively governed by the most immediate determination.
7.2 Causal self-interpretation
The term self-information would suggest an accumulation of data about oneself. The process is more active: the system builds a partial interpretation of the causes that constitute it, and that interpretation itself enters the causal chain. Interpreting a trace does not mean only associating a verbal narrative with it. It may produce a new bodily association, a modification of attention, an acquired inhibition, a new rule of action or a transformation of the environment.
Reflexive consciousness thus resembles an editor applied to its own memory, but an incomplete editor. It possesses neither the full text of its history nor the source code of its functioning. The editor was itself produced by the traces it is trying to modify, and each correction becomes a new inscription that will transform subsequent readings. What is freed is not the blank page, but a limited capacity for editing.
8. Temporal reach: from the immediate past to future generations
Levels of organisation are not distinguished by a simple speed. Biology can act in milliseconds or over a whole life. A cognitive decision may be instantaneous or prepare several years ahead. A cultural transformation may spread within hours or sediment over centuries. The relevant progress concerns the temporal reach rendered causally accessible.
A system gains historical depth when it can preserve a modification longer, reactivate it in different contexts, associate it with more distant future consequences, transmit that relation beyond its own existence and revise the rules of that transmission. Anticipation renders the absent future active in the present. Writing renders an absent past active for other individuals. A constitution can make old conflicts bear on future decisions. A reform can modify the conditions under which that constitution will itself be revised.
CAUSAL REACH Higher levels do not necessarily act more slowly. They render causally accessible more distant portions of the past and the future, and increase the system's capacity to modify the rules of that relating. |
|---|
9. Collective branch: culture as a politics of traces
9.1 Externalising in order to accumulate
Human reflexivity is not produced by an isolated brain. Language, narratives, diagrams, tools, archives and institutions preserve traces beyond individual memory. They render experiences transmissible to people who have not lived them and allow solutions to be compared beyond a single life.
Iterated-transmission experiments conducted on artificial languages show how structures can emerge and be transformed along successive chains of learning [18]. Cumulative culture thus extends causal reach: one individual's solution becomes an initial condition for the next. It is not a passive memory. It modifies the possibilities of perception, reasoning and action of the systems that enter it.
Naming a state as trauma, injustice, discrimination or conditioning does not abolish it. Naming does, however, render certain relations visible, comparable and contestable. The individual can reread their history because they inherit collective tools of reading. Approaches that resituate cognition within bodily and sensorimotor organisation likewise stress that mental categories must not be isolated from the forms of activity and coordination that make them possible [19, 20].
9.2 Inscribing, naming, interpreting, revising
Culture is nevertheless not a neutral reservoir. It is a field of forces in which access to the fundamental operations is unequally distributed. Who decides what deserves a durable inscription? Which categories are recognised for naming an experience? Who has the authority needed to impose an interpretation? Who can modify the rules, the narratives and the institutions that follow from it?
Contemporary digital technologies intensify this politics of traces. Recommendation algorithms select which contents gain visibility and can modify the very structure of collective exchange [27]. Surveillance systems convert behaviours into persistent data, liable to be cross-referenced and reinterpreted. Generative AI systems, trained on vast cultural corpora, now intervene in the reproduction, reformulation and ranking of these inheritances [28, 29]. These technologies are therefore not mere supports: they become powerful actors in the inscription, interpretation and revision of collective traces. Their power nevertheless calls for particular caution: such systems can amplify or rigidify certain traces by reproducing biases, producing plausible but unfounded content, and concentrating the power of selection and interpretation among a limited number of actors.
Collective ethics therefore does not consist merely in adding up individual preferences. It organises the confrontation between different normativities, positions and temporal horizons. An institution may increase a group's memory and coordination while preventing the people affected from revising it. Collective freedom depends less on the quantity of traces accumulated than on the real reciprocity of the power to act upon them.
POLITICS OF TRACES A free society is not merely a society that preserves a great deal of information. It is a society in which the capacities to inscribe, name, interpret, contest and revise common traces are effectively distributed. |
|---|
10. A testable research programme
The framework would remain a philosophical vocabulary if it produced no empirical difference. Its propositions can be turned into predictions and criteria of refutation. Experiments will not test the whole architecture in a single move. They can, however, put to the test the transitions and dissociations on which it rests.
Table 1: predictions, empirical criteria and types of possible experiment.
| Prediction | Empirical criterion / index of refutation | Type of possible experiment |
|---|---|---|
| 1. Causal efficacy of a trace | Selective perturbation of a persistent modification must change a future response in a comparable situation. If no outcome is modified, the qualification as a functional trace is weakened. | Optogenetic or pharmacological perturbation of memory circuits. Targeted epigenetic editing. Inhibition of a sensory adaptation mechanism. Before / after perturbation protocols. |
| 2. Dissociation of valence | Detection, learning or action may be relatively preserved while the reported affective component or the hedonic reactions are modified. | Lesion and clinical studies. Pharmacological manipulation of dopaminergic or opioid systems. Brain stimulation. Joint measurement of behaviour, physiology and reported experience. |
| 3. Metacognitive generalisation | Adjustment of uncertainty, confidence or information seeking must transfer to new tasks and resist explanation by simple local association. | Opt-out and information-seeking paradigms in primates or rodents. Transfer tasks. Variation of reward contingencies. Models comparing metacognitive strategies with first-order learning. |
| 4. Durable reflexive revision | The modification of a criterion must persist, transfer, change the hierarchy among several responses and influence the acquisition of new traces. | Longitudinal studies of cognitive reappraisal or rule learning. Metacognitive interventions. Behavioural and neural measures before, during and after transfer to new contexts. |
| 5. Collective revisability | With comparable memory and resources, groups endowed with real procedures of contestation and correction should better integrate errors and adapt over the long term. | Institutional comparisons and natural experiments. Cultural transmission paradigms. Deliberation laboratories. Multi-agent simulations testing the openness or closure of revision rules. |
11. Objections and limits
11.1 Does the model attribute mental capacities too readily?
The risk of anthropomorphism is real if the words evaluation, preference or memory are used without precision. The framework answers with graded definitions. A regulatory discrimination is not a conscious judgement. A trace is not a recollection. A functional priority is not a feeling. A behavioural simulation is not necessarily a conscious imagination. The same terms must not mask different mechanisms.
11.2 Does functionalism resolve the hard problem?
No. The architecture describes the functional conditions of an increasing access to traces and constraints, but it does not demonstrate why certain states are experienced. It turns the hard problem into an explicit lock and makes it possible to distinguish better what has been explained from what has merely been redescribed. This limitation is constitutive of the programme, not a detail provisionally concealed.
11.3 Is reflexive revision causally impotent?
A strict epiphenomenalism might hold that experience accompanies the mechanisms without producing anything. The framework does not need to assert that qualia, taken in isolation, exert an additional force. It asserts that representations, relations of confidence, verbal rules and self-models are physical and symbolic configurations that modify behaviours and learning. Whether their lived character possesses a distinct causal efficacy remains an open question.
11.4 Does the language of inheritance introduce a cosmic teleology?
No cosmic finality is necessary. The properties of transformation, inscription and reorganisation of matter make possible systems capable of thinking, without thought being prefigured as a goal. The living exploits certain dynamics because organisations able to persist and reproduce were selected, not because the Universe was intentionally preparing the appearance of consciousness.
11.5 Can multi-level coherence become a new dogma?
Multi-level coherence is not a definitive moral calculus. It requires making visible the transfers of cost, the dependencies and the delayed effects. Its own criteria must remain open to discussion. It functions as a procedure of vigilance: widening the field of consequences taken into account and maintaining the possibility of correction, rather than claiming to abolish all conflict of value.
12. Theoretical positioning
The framework belongs to a historical and organisational naturalism. It shares with the theory of biological autonomy the idea that functions and normativity must be referred to the self-maintenance of an organisation [1, 2, 22]. It joins enactive and sensorimotor approaches where they refuse to reduce cognition to an abstract computation separated from body and action [19, 20]. It is compatible with active inference where the latter formalises how systems maintain certain states, update expectations and act on their environment [21].
It is nevertheless not identical with any of these frameworks. Active inference does not demonstrate phenomenal valence. Organisational biology does not by itself supply a theory of consciousness. Neural theories of consciousness do not always integrate cultural externalisation and the politics of traces. The model's own contribution consists in linking these domains through a single question: to what degree can a system access the effects of its history and intervene in the conditions of their reactivation?
This formulation also opens a question for artificial intelligence. An artificial system might have traces, counterfactual models, monitoring of its errors and even rules of revision without exhibiting phenomenal valence. It could therefore reach certain levels of the functional and collective branches without that sufficing to establish a lived experience. The architecture thus avoids confusing reflexive performance with consciousness.
Conclusion: when history begins to act upon itself
Matter perhaps does not contain consciousness in hidden form. It nevertheless possesses a historicity: its transformations can leave effects that orient subsequent transformations. The framework proposed describes how that historicity can be exploited, lived, represented, externalised and partly revised, without supposing that each of these transitions is automatic.
Matter preserves certain effects of its transformations. The living organises them into norms of viability. Cognition mobilises them to select and anticipate. Consciousness integrates them into a situated experience. Reflexivity renders them partly representable and revisable. Culture makes them the object of a politics. Freedom appears when that revision itself enters the chain of causes and modifies how the history continues.
Freedom is therefore neither a blank page nor a magical exception to physical laws. It is the progressive widening of the confrontation between causes, of their temporal reach and of their revisability. It exists when a system no longer merely undergoes being produced by its history, but can use a partial representation of that history to modify the way it will continue to determine it.
We inherit the chaos of the Universe in the sense that we remain constituted by a matter that is fluctuating, sensitive, historical and capable of changing regime. This inheritance does not necessarily remain an agitation merely undergone. Fluctuation can become exploration. Instability, plasticity. The trace, memory. The bifurcation, a represented possibility. Conflict, deliberation. Repetition, an object of understanding.
Consciousness does not organise chaos from outside. It is a provisional organisation of that chaos, become partly able to work on the conditions of its own persistence and its own transformation.
Synthesis of the architecture
| Regime | Principal branch | Capacity added |
|---|---|---|
| Material historicity | All | Transformations leave effects that modify what follows. |
| Biological normativity | Functional | Certain traces are used relative to a domain of viability. |
| Functional valence | Functional | States receive a global priority that organises action and learning. |
| Phenomenal lock | Phenomenal | Functional priority does not by itself allow a lived experience to be deduced. |
| Situated consciousness | Phenomenal | Certain states are integrated into a present, oriented bodily experience. |
| Preference and anticipation | Functional | Traces and values serve to select actions and compare absent trajectories. |
| Reflexivity | Functional and collective | Certain criteria become representable and partly revisable. |
| Multi-level coherence | Transversal | Revision is evaluated according to its relational and temporal consequences. |
| Politics of traces | Collective | The powers to inscribe, name, interpret and revise are externalised and contested. |
References
[1] Montévil, M. & Mossio, M. (2015). Biological organisation as closure of constraints. Journal of Theoretical Biology, 372, 179–191. https://doi.org/10.1016/j.jtbi.2015.02.018
[2] Mossio, M., Saborido, C. & Moreno, A. (2009). An organizational account of biological functions. The British Journal for the Philosophy of Science, 60(4), 813–841. https://doi.org/10.1093/bjps/axp036
[3] Segall, J. E., Block, S. M. & Berg, H. C. (1986). Temporal comparisons in bacterial chemotaxis. Proceedings of the National Academy of Sciences, 83(23), 8987–8991. https://doi.org/10.1073/pnas.83.23.8987
[4] Sourjik, V. & Wingreen, N. S. (2012). Responding to chemical gradients: bacterial chemotaxis. Current Opinion in Cell Biology, 24(2), 262–268. https://doi.org/10.1016/j.ceb.2011.11.008
[5] Schultz, W., Dayan, P. & Montague, P. R. (1997). A neural substrate of prediction and reward. Science, 275(5306), 1593–1599. https://doi.org/10.1126/science.275.5306.1593
[6] Berridge, K. C. & Robinson, T. E. (1998). What is the role of dopamine in reward: hedonic impact, reward learning, or incentive salience? Brain Research Reviews, 28(3), 309–369. https://doi.org/10.1016/S0165-0173(98)00019-8
[7] Daw, N. D., Gershman, S. J., Seymour, B., Dayan, P. & Dolan, R. J. (2011). Model-based influences on humans’ choices and striatal prediction errors. Neuron, 69(6), 1204–1215. https://doi.org/10.1016/j.neuron.2011.02.027
[8] Pfeiffer, B. E. & Foster, D. J. (2013). Hippocampal place-cell sequences depict future paths to remembered goals. Nature, 497, 74–79. https://doi.org/10.1038/nature12112
[9] Berthier, M., Starkstein, S. & Leiguarda, R. (1988). Asymbolia for pain: a sensory-limbic disconnection syndrome. Annals of Neurology, 24(1), 41–49. https://doi.org/10.1002/ana.410240109
[10] Sanz Perl, Y. Et al. (2021). Nonequilibrium brain dynamics as a signature of consciousness. Physical Review E, 104, 014411. https://doi.org/10.1103/PhysRevE.104.014411
[11] Tagliazucchi, E. Et al. (2016). Large-scale signatures of unconsciousness are consistent with a departure from critical dynamics. Journal of the Royal Society Interface, 13, 20151027. https://doi.org/10.1098/rsif.2015.1027
[12] Dehaene, S. & Changeux, J.-P. (2011). Experimental and theoretical approaches to conscious processing. Neuron, 70(2), 200–227. https://doi.org/10.1016/j.neuron.2011.03.018
[13] Tononi, G., Boly, M., Massimini, M. & Koch, C. (2016). Integrated information theory: from consciousness to its physical substrate. Nature Reviews Neuroscience, 17, 450–461. https://doi.org/10.1038/nrn.2016.44
[14] Seth, A. K. & Bayne, T. (2022). Theories of consciousness. Nature Reviews Neuroscience, 23, 439–452. https://doi.org/10.1038/s41583-022-00587-4
[15] Ferrante, O. Et al. / Cogitate Consortium (2025). Adversarial testing of global neuronal workspace and integrated information theories of consciousness. Nature, 642, 133–142. https://doi.org/10.1038/s41586-025-08888-1
[16] Fleming, S. M., Ryu, J., Golfinos, J. G. & Blackmon, K. E. (2014). Domain-specific impairment in metacognitive accuracy following anterior prefrontal lesions. Brain, 137(10), 2811–2822. https://doi.org/10.1093/brain/awu221
[17] Hampton, R. R. (2001). Rhesus monkeys know when they remember. Proceedings of the National Academy of Sciences, 98(9), 5359–5362. https://doi.org/10.1073/pnas.071600998
[18] Kirby, S., Cornish, H. & Smith, K. (2008). Cumulative cultural evolution in the laboratory: an experimental approach to the origins of structure in human language. Proceedings of the National Academy of Sciences, 105(31), 10681–10686. https://doi.org/10.1073/pnas.0707835105
[19] Keijzer, F., van Duijn, M. & Lyon, P. (2013). What nervous systems do: early evolution, input-output, and the skin brain thesis. Adaptive Behavior, 21(2), 67–85. https://doi.org/10.1177/1059712312465330
[20] Keijzer, F. (2015). Moving and sensing without input and output: early nervous systems and the origins of the animal sensorimotor organization. Biology & Philosophy, 30, 311–331. https://doi.org/10.1007/s10539-015-9483-1
[21] Friston, K. (2013). Active inference and free energy. Behavioral and Brain Sciences, 36(3), 212–213. https://doi.org/10.1017/S0140525X12002142
[22] Moreno, A. & Mossio, M. (2015). Biological Autonomy: A Philosophical and Theoretical Enquiry. Dordrecht: Springer. https://doi.org/10.1007/978-94-017-9837-2
[23] Hampton, R. R. (2019). Monkey metacognition could generate more insight. Animal Behavior and Cognition, 6(4), 230–235. Https://doi.org/10.26451/abc.06.04.02.2019
[24] Templer, V. L. (2019). Slow progress with the most widely used animal model: Ten years of metacognition research in rats, 2009–2019. Animal Behavior and Cognition, 6(4), 273–277. Https://doi.org/10.26451/abc.06.04.07.2019
[25] Cocchi, L., Gollo, L. L., Zalesky, A. & Breakspear, M. (2017). Criticality in the brain: A synthesis of neurobiology, models and cognition. Progress in Neurobiology, 158, 132–152. Https://doi.org/10.1016/j.pneurobio.2017.07.002
[26] Breakspear, M. (2017). Dynamic models of large-scale brain activity. Nature Neuroscience, 20, 340–352. Https://doi.org/10.1038/nn.4497
[27] Santos, F. P., Lelkes, Y. & Levin, S. A. (2021). Link recommendation algorithms and dynamics of polarization in online social networks. Proceedings of the National Academy of Sciences, 118(50), e2102141118. Https://doi.org/10.1073/pnas.2102141118
[28] UNESCO. (2021). Recommendation on the Ethics of Artificial Intelligence. Paris: UNESCO. https://unesdoc.unesco.org/ark:/48223/pf0000381137
[29] UNESCO Independent Expert Group on Artificial Intelligence and Culture. (2025). Artificial Intelligence and Culture. Paris: UNESCO.
Document prepared within the ORI-C programme: ori-c.be