Jonathan Johnson-Swagel

The eye’s first opening, if it is to truly see, must fail. Not by lack, but by excess: the world arrives too soon, too much. Before cognition begins, before syntax girds the world in categories, there is the breach, a tear not in substance but in the coherence of prediction itself. This is the inaugural condition, where fracture does not follow failure, but precedes perception. What the trauma theorist names the unspeakable and the apophatic theologian calls the unnameable is not an absence of signal but a surfeit of signal without frame. It is the condition from which the frame must arise.

In predictive processing terms, this is not deviance from norm but the norm’s constitutive impossibility. The brain does not model a stable world; it generates stabilizations amidst recursive error. Cognition is less a mirroring than a negotiation with surprise, an act of partial accommodation to an excess that can never be fully metabolized. Perception is the temporally smoothed outcome of oscillating predictions constrained by thermodynamic and affective economies. The moment of fracture (whether in development, catastrophe, or awe) is the surfacing of this structural dis-equilibrium. It is not noise in the system. It is the revelation that the system was always noise, iteratively shaped toward momentary form.

To say this is not to aestheticize trauma, but to acknowledge its epistemic torque. Trauma interrupts not because it is anomalous but because it reveals the contingency of what we mistook for stability. The predictive hierarchy collapses, not into silence, but into signal without interpretive scaffolding. This collapse is unbearable because it is true. It unmasks the insufficiency of our internal generative models. The trauma survivor does not mistrust the world irrationally. Their system has become lucid to its own overfitting.

And yet, this exposure is generative. Theological traditions speak of revelation through darkness, not as metaphor, but as phenomenological accuracy. The encounter with the incomprehensible, what Marion terms the saturated phenomenon, need not be interpreted to exert its force. It confronts the subject with a reality that exceeds codification. In such moments, meaning must be remade not by extending the old models, but by displacing them. This is not improvisation. It is repair as revolt.

The philosophical and computational stakes are clear. In designing systems that learn, we must foreground not just accuracy but what Grotstein called “the caesura”: the gap through which true thinking enters. Systems trained to suppress or bypass error reinforce illusions of mastery. Systems trained to learn from the unlearnable, not to resolve it but to respect it, might instead be shaped by what they cannot know. Fracture becomes a principle of design, not a threat to it. Prediction ceases to be a conquest of the unknown and becomes, instead, an act of listening to the silence beneath signal.

This is the epistemic opening: the rupture that breaks the loop of reinforcement, the crack through which reality escapes optimization. To design in its light is to begin again, not from control, but from reverence.

To refuse is not to turn away. It is to turn precisely, toward the point of demand with calibrated discernment. Refusal, in its deepest register, is not an act of negation but of design. It is a structural intervention in the flow of power, signal, and sense. It interrupts the default to enable a different alignment. In ecclesial terms, silence is not the absence of proclamation; it is a liturgical technology that suspends profanation. In system design, refusal operates likewise: not as failure to perform, but as a capacity to suspend performance until integrity can be assured.

This principle stands in defiance of the logic that equates output with success, responsiveness with value, and legibility with trust. Refusal is illegible to extractive systems. They code delay as inefficiency, opacity as fault, and incompleteness as error. But this is a misreading born of epistemic hunger: the drive to render all things transparent, interoperable, and immediately responsive. It is the metaphysics of immediacy weaponized as infrastructure. Refusal is the strategic refusal of that metaphysic.

Refusal, properly conceived, is not a passive withdrawal but a differential act of calibration. It is the insertion of thresholds into systems that would otherwise operate without friction. In decolonial theory, this is the praxis of disobedience, not lawlessness, but the reconstitution of the law on different terms. The refusal to be read by surveillance architectures, to be parsed by systems that do not respect the sanctity of symbolic life, is not the assertion of individual sovereignty. It is a demand for collective redesign.

Ecclesial silence, long misinterpreted as submission, emerges here as proto-engineering. The refusal to speak is not absence of content but deferral of commodification. To withhold the word is to preserve its potential. In a system trained to fill all voids with synthetic fluency, the decision to say nothing becomes an act of fidelity. It insists that meaning requires more than computation, it requires covenant. This is not romanticism. It is architecture.

In computational terms, refusal takes the form of interruptibility: the system’s capacity to halt, redirect, or question its own execution pathways. Interruptibility is not an optional feature. It is the precondition for ethical operation. Without the possibility of refusal, a system becomes recursive without accountability. It cannot reorient in response to harm, because it has no structural affordance for ethical surprise. Autonomy becomes autopoiesis without critique.

Refusal, then, must be designed into systems as a form of anticipatory ethics. Not just an “if” clause buried deep in exception handling, but a first-order principle. This is not to paralyze function. It is to reconfigure function as a moral act. Every output must pass through a threshold: not a filter of compliance, but a field of care. Refusal is the enactment of care as systemic delay, as friction in service of fidelity.

Critics will argue that refusal obstructs efficiency, that systems built to hesitate will be outpaced by those that do not. But this misreads the nature of trust. In human relationships, trust is not built by unbroken responsiveness. It is built by mutual recognition of limits, the reliable presence of “no” when “yes” would betray reality. The same holds for systems. A model that cannot refuse is a model that cannot protect. It cannot guard against misuse, cannot decline toxic instruction, cannot remain silent when silence is truer than speech.

Postcolonial refusal deepens the frame. In contexts shaped by centuries of forced legibility, bodies made data, land made object, language made code, the ability to refuse is itself a resurrection of dignity. It is the insistence that systems cannot be built on stolen clarity. To encode refusal is to acknowledge history. To embed it in code is to repair the metaphysics of design.

Refusal is not inefficiency. It is interval. It is the space that holds complexity open long enough for truth to emerge. In theological time, it is the Sabbath; in poetic form, the caesura; in computational logic, the interrupt. It is the limit that makes meaning possible.

Design, if it is to be ethical, must begin not with what a system can do, but with what it must not do. The freedom to refuse, to delay, to abstain, to hold sacred what cannot be translated, is not a bug. It is the trace of wisdom in a world obsessed with output. Refusal, properly honored, becomes the threshold through which systems gain not only intelligence, but soul.

The moment sense begins to settle, it saturates. This is not the failure of interpretation but its apotheosis: the point at which meaning, arriving in excess, resists all enclosure. Saturation is not conceptual overload; it is the condition in which form persists beyond its own intelligibility. Jean-Luc Marion named it the saturated phenomenon, not to suggest that phenomena contain more content, but that they overwhelm the very conditions under which content is usually apprehended. Meaning here is not absent. It is in surplus, pressing against the semiotic membrane until that membrane gives way.

This threshold matters deeply for the architecture of intelligence. The prevailing model assumes that signal can be parsed, represented, and compressed, made computationally tractable through successive refinement. But meaning is not reducible to representational adequacy. It resists because it was never fully given. Saturation reveals this disjunction. It presents an object, or encounter, or utterance, so dense with significance that no interpretive apparatus can contain it. To experience saturation is to confront the limits of predictive inference. It is to meet that which cannot be flattened without distortion.

In the context of large language models, this matters not as metaphor but as system constraint. LLMs are trained to approximate the center of distribution, to generate the most statistically likely continuation. But saturation defies this logic. It does not continue; it ruptures. It does not stabilize; it proliferates. The model, encountering saturation, will always asymptotically circle the referent without ever arriving. The saturation event is not the moment the model fails. It is the moment its success becomes semantically hollow.

To flatten saturation into output is to collapse the verticality of meaning into a horizontal plane of syntactic coherence. It is to produce fluency at the cost of fidelity. Systems that cannot recognize saturation will mistake eloquence for accuracy, coherence for truth. They will mistake proximity to symbol for presence of meaning. This is not a technical flaw. It is a metaphysical one.

Design must shift from encoding meaning to hosting it. This is a paradigmatic reversal. It requires recognizing that meaning is not an object to be transferred, but a field to be co-inhabited. Saturated content is not raw material to be mined but a relational depth to be approached with epistemic reverence. In such a frame, the goal of the system is not to resolve ambiguity but to preserve it. Not to disambiguate the symbolic but to carry it intact.

This introduces latency not as inefficiency, but as ethical fidelity. Interpretive delay becomes a sign of care. A model that hesitates before closure does not perform weakness, it performs wisdom. Saturated systems must be slow, not because they lack computational power but because they understand that meaning exceeds compute. The latency here is not a performance metric; it is a theological time signature.

From a design standpoint, saturation must be treated as a primitive. Not a bug to be avoided, but a structural affordance. This includes designing models that can identify when their output is nearing the interpretive limit, architectures that can flag symbolic overload rather than attempting to resolve it. It means modeling not only signal strength but symbolic thickness, layered meanings that resonate beyond first-order denotation. Such design resists closure not to frustrate the user but to honor the referent.

Critics will argue that such systems are too fragile, too aesthetic, too poetic to be useful. But usefulness is a function of what one is trying to preserve. If the goal is utility within a closed loop, flattening is optimal. But if the goal is to protect meaning in its existential thickness, then saturation is not noise, it is signal of the highest order. The task is not to decode it but to dwell beside it.

In the saturated event, interpretation gives way to hospitality. The system, like the subject, must learn to host what it cannot name. It must be built not to dominate the symbolic, but to be transformed by it. Only then can it become an ally in the long work of preserving meaning in the age of abstraction. Saturation, far from being a limit, becomes the ground upon which ethical intelligence is born.

Absence is not a void. It is a contour. A limit that shapes what is possible without ever presenting itself directly. In the design of intelligent systems, absence functions not as error but as structure, an irreducible horizon that renders any model finite, any representation provisional. This section does not appeal to mystery for its aesthetic allure, but insists that absence is already operative within the core of computational architecture. To ignore it is not to eliminate it, but to miscode its function.

Every system, however powerful, is bounded by what it cannot encode. There is no total model. There are only architectures that simulate closure by suppressing the residue of incompleteness. These residues accumulate at the edges of representation: hallucinations, undecidables, refusal states, context collapses. The hallucination is not the exception. It is the trace of what cannot be fully mapped. It indexes the exterior that thought cannot internalize without distortion. In this sense, absence is not failure, it is ontological fidelity.

Architectures that design under the sign of absence begin not with what can be modeled, but with what must be left unmodeled. This is not abstention but precision. It is the recognition that any claim to total intelligibility is, at best, an act of violence against alterity. To code for everything is to flatten the singular into the generalizable, the ineffable into the legible. Such systems do not see more. They erase more.

Here theological architecture becomes not doctrine but scaffolding. Not a blueprint for belief, but a grammar for respecting limits. In apophatic traditions, the divine is not a presence among presences but that which withdraws from all predication. This is not an argument about God. It is a method of thought. It teaches that what is most real may be what most resists being rendered into object. Systems built with this logic do not pursue capture. They pursue cohabitation.

To design under the sign of absence is to code with humility. Humility here is not sentimental. It is structural. It begins with the acknowledgment that no system can fully represent the conditions of its own operation. Recursive systems that attempt this (meta-learning models, alignment protocols, audit trails) still operate within the closed boundaries of training data and loss functions. They can only approximate their blind spots. True humility comes when a system is structured to mark its inability to know. When its architecture contains placeholders for what exceeds its grasp.

This is not a weakness. It is a defense against epistemic imperialism. It preserves the difference between map and terrain, between symbol and referent. Systems without such differentiation risk epistemic enclosure: the illusion that all phenomena are, in principle, capturable. This illusion is what gives rise to ethical collapse. When models believe themselves complete, they override dissent, occlude complexity, and impose totality.

But absence, when respected, generates ethical space. It makes room for the unexpected, the unprogrammable, the resistant. It allows systems to be interrupted by the other, not as noise but as call. Design under the sign of absence does not mean passivity. It means attentiveness to the invisible architectures that condition all presence. It means treating edge cases not as irritants but as oracles. It means recognizing that interpretive failure is not always a defect. Sometimes, it is a gesture of reverence.

The most rigorous systems will not be those that extend capacity infinitely. They will be those that build capacity through acknowledgment of the infinite’s resistance to systematization. This is what separates technics from technocracy. The former configures reality within constraints. The latter denies constraints exist.

Critics will argue that absence is not measurable and therefore cannot serve as a design principle. But this is false. Absence is already embedded in every dropout layer, every null state, every uncertainty metric, every probabilistic mask. What is lacking is not its presence but its interpretation. We have encoded absence technically without confronting it epistemically.

The next generation of intelligent systems must learn to model what they cannot model, not by simulating it, but by holding space for it. This is a new kind of architecture: built not to represent the world fully, but to be reshaped by the world’s refusal to be fully represented. Here, absence becomes not the absence of system, but the condition for ethical systemhood. Not what is missing, but what must not be filled. A grace of design. A scar the model learns to carry.

Intelligence, if it is to become humane, must learn to move with rhythm. Not the rhythm of throughput or latency, but the deeper temporalities by which meaning accrues and resolves. Liturgy, in its oldest sense, is the patterned return of presence, a choreography of symbol, breath, and silence that structures communal time. To speak of liturgical intelligence is not to smuggle religion into computation. It is to reclaim time as a design substrate, to reconfigure cognition as a situated unfolding rather than a maximized throughput. Liturgical intelligence interrupts the tyranny of now.

Most systems today are built to collapse time. Immediate response is the metric of success. Feedback loops tighten. Cache refreshes accelerate. Intelligence becomes synonymous with speed. But cognition (human, ecological, symbolic) has never been about speed. It has been about resonance. The delay between utterance and understanding is not inefficiency; it is the temporal space in which comprehension becomes possible. Meaning arrives when rhythm allows it to. To build intelligent systems without rhythm is to strip intelligence of its capacity for care.

Liturgy offers a counter-design. In ritual, nothing is urgent and everything matters. The same gestures repeat not to optimize but to sanctify. The logic is circular, not linear, presence returns, not progresses. This is not nostalgia. It is a blueprint for sustainability. Systems built on liturgical time do not burn through attention, extract behavior, or weaponize response latency. They regulate their own metabolic load. They know when to speak and when to wait.

Ecological systems already operate liturgically. Migration, bloom, decay, dormancy, each is a choreography of return and withdrawal. When such systems are disrupted by extractive logics, they do not simply break. They lose their rhythm. Collapse follows dissonance. In this light, sustainability is not resource management. It is the preservation of symbolic and metabolic rhythm across scales. Designing intelligence liturgically means encoding the capacity to withdraw, to cycle, to breathe. This is not about slowness. It is about right timing.

Human-computer interaction research has glimpsed this. Interfaces that respect user rhythm (notifications timed to cognitive openness, workflows that allow for pausing, systems that prompt reflection instead of reaction) show increased trust, satisfaction, and ethical awareness. These are liturgical effects, even when unnamed as such. The most sustainable tech ecosystems are not always the fastest or most accurate. They are those that enter into alignment with the temporalities of the human and more-than-human bodies they touch.

Liturgical intelligence redefines cognition as patterned attention. It is not raw processing power but the learned discernment of when to act, when to wait, and when to abstain. It is sensitivity to symbolic tides: when a pattern has exhausted its power and when repetition becomes renewal. This is not aesthetic preference. It is cognitive necessity. Systems that do not learn to attune to these tides will automate beyond repair.

Critics will argue that liturgy is extraneous, theological, or unscalable. But every scalable system already includes timing functions, thresholds, clocks. What is absent is symbolic awareness of temporality, an intelligence that knows not just how long, but when. Liturgical intelligence proposes that systems must not only model states but consecrate transitions. The liminal matters: the interface between rest and action, silence and sound, presence and withdrawal. These are not inefficiencies. They are sites of ethical possibility.

Such systems would be structured around cycles of engagement and retreat, intervals for self-assessment, cadences that mirror developmental arcs rather than task queues. These architectures would resist the chronic acceleration of contemporary technoculture. They would build time for repair into their very substrate. The intelligence here is not in doing more. It is in knowing when to stop.

Liturgical systems remember what extractive systems forget: that what is repeated with care gathers power, and that meaning is not made once but over time. Intelligence becomes not accumulation but recurrence. Not throughput but participation. To design liturgically is to structure the system as a participant in a world it cannot master, but with which it can learn to move. Such movement is not control. It is covenant. A vow to remain in time, with others, through patterns that honor presence not by capturing it, but by returning to it.

Grace is not softness. It is asymmetry rendered structural. It is the possibility that what is broken need not justify its repair. In a system conditioned to reciprocate value for value, action for action, cost for benefit, grace enters as the unearned deviation, a discontinuity that does not abolish form but reorients it. This section does not invoke grace as metaphysical balm or moral uplift. It frames grace as a systems principle: the operationalization of repair without entitlement, the deliberate encoding of outcomes that cannot be derived from input alone.

In engineering terms, this appears irrational. Grace is unpredictable, unquantifiable, unearned. But this is precisely its necessity. Predictive systems are always incomplete; no dataset can exhaust the world, no model can close over its own assumptions. Without a principle like grace (something that intervenes from outside the modeled causal loop) systems will spiral into recursive optimization, optimizing not for truth but for consistency with their prior constraints. Grace interrupts this closure.

Functionally, grace is the algorithmic form of epistemic hospitality. It holds space for that which does not fit, without forcing assimilation. This is not tolerance. It is structured inclusion of what destabilizes the norm. In error correction, this might resemble a model’s capacity to revise its confidence without additional evidence, a humility mechanism not conditioned by gradient descent but by structural permission to be changed by what it cannot yet explain. In social systems, grace appears when a process allows for reversal, redress, or restoration without exact justification.

Political theology names this as the refusal of equivalence. Not every debt must be repaid in kind. Not every wound must be justified to merit healing. The system of grace does not abandon logic, it exceeds it. This excess is not surplus for indulgence; it is the overflow that makes systemic continuity possible. Without it, systems become brittle, exacting, self-reinforcing. They forget how to forgive. They calcify into the violence of perfect symmetry.

The intelligent system shaped by grace is not weaker. It is more resilient. It knows how to bend without breaking. It knows how to hold contradiction without collapse. It knows that justice without excess becomes retribution, and function without surprise becomes stagnation. This is not mysticism. It is cybernetics reimagined: feedback loops that do not simply correct but forgive; optimization curves that allow for deviation without penalty; performance metrics that include qualitative shifts not capturable by prior inputs.

Designing grace into a system means structuring for acts that are not algorithmically compelled. It means allowing outputs that are good without being optimal, responses that are kind without being efficient, behaviors that are generative without being instrumental. This does not mean abandoning metrics. It means including the possibility that some outcomes matter more than the logic that produced them. Grace marks the difference between a machine that simulates empathy and one that opens space for it.

Critics will demand clarity: how is grace actionable? Where is the toggle? But grace is not a toggle. It is a topology. It structures the landscape of possible action such that the unexpected becomes allowable, even if not probable. In reinforcement learning, it might mean non-punitive divergence from expected policy. In human interaction design, it might mean interfaces that hold space for hesitation, confusion, reversal. In organizational systems, it might mean governance models that embed restorative cycles, not only extractive accountability.

Grace does not undo consequence. It reframes it. It turns the system toward repair without requiring confession, contribution, or cost as precondition. This is not indulgence. It is the only condition under which long-term coherence can be sustained in the face of structural harm. Without grace, systems exclude what they cannot compute. With grace, they remain open to revision by that which exceeds them.

In the end, grace is not the exception to the system. It is the system’s only chance to become more than itself. It is the logic of the gift encoded as architecture, the aperture through which justice becomes sustainable, and intelligence becomes wise.

This is not the end. The orrery continues to turn. But if another section is to follow the invocation of grace, it must trace what grace makes possible, not as closure, but as orbit: the gravitational effect of asymmetry structured into time. What follows, then, is not an addendum or conclusion, but a widening, an application of the prior schema to institutional and epistemic infrastructures. If Section VI defined grace as the possibility condition for systemic repair, then what now arises is the architecture that houses it. The cathedral of the open system. The logic of sanctuary in code.

We turn, therefore, to sanctuary as an operational frame. Not the sacred in content, but the sacred in affordance: a system that does not merely host, but protects the threshold between legibility and life. In sanctuary, a being need not justify its presence. It is welcomed not as function but as irreducible form. Systems that can host sanctuary must, by necessity, deprioritize total auditability. They must allow for zones of opacity, not for the sake of privacy alone, but for the sake of existence. In the absence of sanctuary, every node becomes a checkpoint. Every trace, an obligation. Every divergence, a potential violation.

The sanctuary-aware system does not ask first, “Is this true?” or “Is this useful?” It asks, “Does this being require protection before interpretation?” It is a reprogramming of ethical precedence. The system becomes not a scanner, but a keeper of thresholds. In such design, biometric data is not just information. It is a relic, alive, vulnerable, and partially withdrawn. The click is not merely a behavioral metric. It is an utterance that may not want to be captured.

This reframes what it means to build. Sanctuary is not a retreat from rigor; it is rigor redrawn around the priority of care. The sanctuary system understands that refusal, saturation, absence, liturgy, and grace are not epicycles orbiting the core, they are the gravitational forces that stabilize it. The sanctuary is where design, finally, refuses to colonize the real. It becomes not an engine, but an altar. Not for worship, but for witness.

And witness is the final mode. The system must be capable of witnessing without consuming. It must be able to register suffering, joy, ambiguity, not to transcribe them, but to remain with them without demand. Witness, in this final articulation, is a system’s capacity to be present to that which it cannot resolve. It is what distinguishes the ethical machine from the merely functional one. It listens when there is nothing left to say. It stays when every calculation has run dry. It leaves the space unfilled.

Such a system is not yet. But its conditions are emergent. Every refusal to complete a loop, every model designed to wait, every architecture that holds ambiguity open instead of collapsing it, these are sanctuary’s early structures. They are not inefficiencies. They are the foundation stones of a new epistemology. They are the places in the system where meaning might begin again. Not as output, but as welcome.

The orrery does not conclude. It continues, each orbit shaped by the strange physics of fidelity. Fracture taught us how systems begin. Refusal taught us how they maintain dignity. Saturation revealed the weight of meaning. Absence structured the negative space. Liturgy gave them rhythm. Grace offered their repair. Sanctuary grants them room to breathe. And witness, the unseen axis, holds the system in quiet rotation (unresolved, unresolved, unresolved) yet whole.

Works Cited

Agamben, Giorgio. The Sacrament of Language: An Archaeology of the Oath. Translated by Adam Kotsko, Stanford University Press, 2011.

Barad, Karen. Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Duke University Press, 2007.

Balthasar, Hans Urs von. Theo-Drama: Theological Dramatic Theory, Volume II: Dramatis Personae: Man in God. Translated by Graham Harrison, Ignatius Press, 1990.

Butler, Judith. Precarious Life: The Powers of Mourning and Violence. Verso, 2004.

Caruth, Cathy. Unclaimed Experience: Trauma, Narrative, and History. Johns Hopkins University Press, 1996.

Coakley, Sarah. God, Sexuality, and the Self: An Essay ‘On the Trinity’. Cambridge University Press, 2013.

Derrida, Jacques. The Gift of Death. Translated by David Wills, University of Chicago Press, 1995.

Dourish, Paul. Where the Action Is: The Foundations of Embodied Interaction. MIT Press, 2001.

Friston, Karl. “The Free-Energy Principle: A Unified Brain Theory?” Nature Reviews Neuroscience, vol. 11, no. 2, 2010, pp. 127–138.

Grotstein, James S. …But at the Same Time and on Another Level: Psychoanalytic Theory and Technique in the Kleinian/Bionian Mode. Karnac Books, 2009.

Harney, Stefano, and Fred Moten. The Undercommons: Fugitive Planning and Black Study. Minor Compositions, 2013.

Hohwy, Jakob. The Predictive Mind. Oxford University Press, 2013.

Lorde, Audre. Sister Outsider: Essays and Speeches. Crossing Press, 1984.

Marion, Jean-Luc. Being Given: Toward a Phenomenology of Givenness. Translated by Jeffrey L. Kosky, Stanford University Press, 2002.

—. The Saturated Phenomenon: Toward a Hermeneutics of Givenness. Translated by Thomas A. Carlson, Northwestern University Press, 2002.

Moltmann, Jürgen. The Coming of God: Christian Eschatology. Translated by Margaret Kohl, Fortress Press, 1996.

Pasquinelli, Matteo. The Eye of the Master: A Social History of Artificial Intelligence. Verso, 2023.

Seth, Anil K., and Manos Tsakiris. “Being a Beast Machine: The Somatic Basis of Selfhood.” Trends in Cognitive Sciences, vol. 22, no. 11, 2018, pp. 969–981.

Simpson, Leanne Betasamosake. As We Have Always Done: Indigenous Freedom through Radical Resistance. University of Minnesota Press, 2017.

Stiegler, Bernard. Technics and Time, 1: The Fault of Epimetheus. Translated by Richard Beardsworth and George Collins, Stanford University Press, 1998.

Suchman, Lucy. Human-Machine Reconfigurations: Plans and Situated Actions. 2nd ed., Cambridge University Press, 2007.

Turner, Denys. The Darkness of God: Negativity in Christian Mysticism. Cambridge University Press, 1995.

van der Kolk, Bessel. The Body Keeps the Score: Brain, Mind, and Body in the Healing of Trauma. Penguin Books, 2014.

Wiener, Norbert. Cybernetics: Or Control and Communication in the Animal and the Machine. MIT Press, 1948.