II. TECHNICS AND THE COMPOSITION OF TIME

Before we can talk about platforms and feeds, we need to slow down and ask a question that sounds almost embarrassingly basic: How does time appear at all? Bernard Stiegler’s wager in Technics and Time is that you cannot answer this question by looking “inside” the subject alone. Any analysis of temporal experience that stays within the head leaves out the technical supports that make human time what it is (Stiegler, 1998). Put more bluntly: there is no human temporality without gadgets.

Stiegler starts from a Husserlian intuition. In classical phenomenology, the flow of consciousness is structured by three intertwined moments: primal impression (what is happening “now”), retention (the just-past that is still held in awareness), and protention (the just-about-to-happen anticipated on the basis of what has come before) (Husserl, 1991). Listening to a melody is the standard example. You do not hear isolated tones; you hear a phrase. The note that is striking now is heard as part of what has just elapsed and what you expect to come next. If retention collapses, the melody disintegrates into meaningless pings.

Stiegler’s key move is to say: that is only half the story. Husserl’s scheme stays at the level of inner time-consciousness. It describes how the subject synthesizes a sequence given to it, but not how the very sequences available to consciousness are produced, stored, and transmitted. In actual human life, we do not only retain internally; we inscribe, record, delegate, and outsource memory into objects. We do not just remember; we write, carve, film, code. And these exteriorizations do not sit neutrally beside “real” consciousness. They fold back into the way experience itself is braided.

To name this, Stiegler distinguishes three kinds of retention:

Primary retention: the immediate holding-together of a just-past within the now. The lingering of the previous note in a melody as the next note sounds.

Secondary retention: the re-activation of something once experienced—remembering a melody you heard last week and humming it to yourself.

Tertiary retention: the exteriorized traces that support and condition experience—scores, recordings, notations, images, texts, calendars, playlists, archives (Stiegler, 1998).

This third level is Stiegler’s real target. Tertiary retentions are not merely backups of the first two. They actively shape what can appear in primary and secondary retention at all. A musical culture that relies on oral transmission has a different relationship to melody and memory than one saturated with recordings and streaming services. A community that records lineage in oral genealogies inhabits ancestry differently from one that stores it in digital databases and DNA kits. What the subject “can” remember or anticipate is already preformatted by the technical milieu in which it moves.

Here we get the first bridge to Gestalt Logos. If Gestalt psychology taught that perception organizes itself into figure and ground—that we do not see raw stimuli but structured wholes—Stiegler is effectively saying: the figure-ground of time itself is technically conditioned. Primary and secondary retentions are the “figure” of lived experience, but tertiary retention is the shifting “ground” upon which those figures arise. When the ground changes—when the organizing milieu of tertiary retention is re-engineered—what counts as a coherent melody, a plausible narrative, or a livable life-course can alter without any single subject deciding so.

Stiegler insists this is not a new problem. The invention of writing already transformed the structure of memory and inheritance; the printing press reorganized the circulation of texts and the formation of publics; cinema and television reconfigured collective experience of history and spectacle. What is distinctive about the contemporary moment is the industrialization and automatization of tertiary retention. No longer just discrete artifacts (a book, a film, a record), tertiary retention now takes the form of vast, networked technical systems that continuously collect, store, and recombine traces of behavior. What once looked like libraries now looks like data centers.

This is where his vocabulary of proletarianization enters. In Stiegler’s earlier work on industrial labor, proletarianization names the loss of know-how when workers are separated from the tools and routines through which they developed practical intelligence. In the age of platforms, something similar happens at the level of temporal know-how. We risk losing not just skills in the narrow sense, but savoir-vivre—the embodied competence of pacing one’s own life, organizing attention, and inhabiting shared rhythms (Stiegler, 1998). When tertiary retention is centrally programmed and remotely updated, the time-structures through which we live become less a matter of collective, contested construction and more a matter of subscription.

Crucially, none of this is framed as a simple “technology is bad” lament. Stiegler repeatedly describes technics as pharmakon: both poison and remedy. Without tertiary retention there is no culture, no law, no science, no scripture, no shared history. Every practice of care, education, and transmission depends on exteriorized supports. The problem is not that memory is outside of us, but that industrial control over those outsides can undermine our capacity to appropriate them as our own. The pharmacological question is therefore: under what conditions can tertiary retentions function as supports for individuation rather than instruments of disindividuation? (Stiegler, 2010).

For Gestalt Logos, this is the hinge. If we take Stiegler seriously, then the fights over “screen time,” AI text, and doomscrolling are not primarily about distraction versus focus or “natural” versus “artificial” life. They are about who configures the tertiary retentions through which time becomes meaningful at all. Architectures of memory are architectures of temporality. To redesign them is to redesign what kinds of Gestalten—perceptual, narrative, political—can crystallize.

Part I, then, leaves us with three working claims:

1. Human temporality is not an inner property of subjects but a relation between retentional flows and technical supports.
2. Tertiary retention is not a passive archive but an active milieu that shapes what can be retained and anticipated in the first place.
3. The contemporary industrialization of tertiary retention is a pharmacological event: it can either erode or sustain our capacity to author our own temporal arcs.

The rest of the article will treat the platform ecosystem as a concrete instantiation of this retentional regime. Where Stiegler writes about cinema and television, we will write about feeds, notifications, and AI-saturated interfaces. Where he speaks of proletarianization of savoir-faire and savoir-vivre, we will translate this into the language of attention metrics, recommendation architectures, and the slow outsourcing of temporal judgment to systems optimized for “engagement.” But the underlying question remains Stiegler’s: through which technical forms does time appear—and who gets to shape them?

III.

PLATFORMS AS TIME-FACTORIES

If Part I stayed largely at the level of conceptual scaffolding—retentions, technics, temporal composition—Part II steps into the concrete: platforms, feeds, and AI-augmented interfaces as factories for time. The claim is not that our devices simply “steal attention.” It is that they engineer retentional environments in which the very shape of what can count as a present, a past, or a future is progressively reformatted.

1. From archives to live infrastructures

Classical tertiary retentions—the book, the film reel, the vinyl record—were discrete artifacts. They could certainly be industrially produced and mass-distributed, but once pressed or printed, they largely stayed put. You encountered them at the pace of your own reading, viewing, or listening. Their temporality was intensive (how deeply you engaged), not yet fully programmable.

Platformized tertiary retention is different. Social media feeds, streaming dashboards, push notifications, and “For You” pages constitute a live infrastructure that is constantly updating in response to aggregated behavior. They don’t simply store traces; they re-rank, re-surface, and re-compose them on the fly. The archive becomes a fluid reservoir that can be tapped and reshaped at any moment according to engagement metrics, A/B tests, and predictive models (Zuboff, 2019).

In Stiegler’s vocabulary, we are dealing with a new phase of industrialization in which tertiary retentions are no longer just manufactured objects but continuously optimized environments. They form the shifting ground on which primary and secondary retention—what I am attending to now, and what I find myself recalling later—can operate at all (Stiegler, 1998).

2. Recommendation as retentional engineering

Consider the basic logic of a recommendation system. It takes past traces of behavior—clicks, watch time, pauses, replays, scroll velocity, dwell time—and converts them into statistical expectations: users who did X are likely to do Y. These expectations then re-enter the environment as ranked lists, auto-play queues, suggested replies, and algorithmically composed timelines.

From a Stieglerian angle, this is not just prediction; it is retentional engineering. Tertiary retention (stored interaction data) is used to shape what appears as a plausible next object of attention. Protention—the felt sense of “what’s coming next”—is quietly outsourced to the recommender. When a platform auto-loads the next video, composes a queue, or fills a blank input field with suggested text, it effectively pre-fills protention on the basis of aggregated secondary retentions.

The point is subtle but important. Husserl describes protention as a horizonal openness structured by past experience (Husserl, 1991). Under platform conditions, that horizon is no longer just the subject’s own sedimented history; it is a computed horizon assembled from the behavioral traces of millions. What you find yourself expecting is partially what has already been seen to keep “people like you” engaged.

3. Design patterns as micro-temporal regimes

These dynamics are not only algorithmic; they are also formalized in interface design patterns:

• Infinite scroll removes natural stopping points, making the “end” of the feed indistinguishable from its beginning. Time is flattened into a continuous present that is always slightly replenished.

• Stories and streaks impose 24-hour or similar windows of visibility, producing micro-deadlines that bind users into recurrent check-in cycles. Miss the window, and a piece of the social world silently expires.

• Read receipts, typing indicators, and “seen at” stamps transform asynchronous communication into something closer to live performance. The past (who responded, who left you on read) is continually folded into the present as a source of anticipatory anxiety and obligation.

Each of these patterns is a way of scripting primary retention (what remains in view), secondary retention (what lingers in memory as salient), and protention (what one fears or hopes will happen next). They are not neutral UX “choices”; they are micro-temporal regimes. They decide where the cuts in time fall—where episodes can begin, where they can end, and when something is experienced as lost, archived, or still pending.

In Gestalt terms, they define default temporal Gestalten: the typical shapes of an evening’s scrolling, a day’s notification cycles, a week’s accumulation of unread messages. Over time, these shapes can become more familiar than the slower rhythms of seasons, liturgical calendars, or life stages.

4. Attention as an industrial input

Yves Citton’s argument about an “ecology of attention” is useful here. Attention, he suggests, is not a purely individual resource but an environmental condition—something collectively shaped by institutions, media, and infrastructures (Citton, 2017). Platforms intensify this by treating attention as an input to be captured, modulated, and sold. The feed is both factory floor and finished product.

This industrial logic feeds directly back into tertiary retention. What gets recorded, highlighted, resurfaced, or buried is keyed to attention metrics: click-through rates, watch time, “session length,” and so on. When a platform tweaks its ranking formula to favor short-form video over links, or comments over original posts, it’s not just changing what people see; it’s changing which traces will count as memorable and therefore which experiences can be easily reactivated later.

The result is a feedback loop:

1. Behavior leaves traces (tertiary retention).
2. Traces are processed into predictions and rankings.
3. Rankings shape what appears as the next “now.”
4. That “now” produces new traces, which refine the model.

Over time, this loop can narrow the range of lived temporal possibilities: what you see, what you recall, and what you expect are increasingly drawn from a subset of patterns already optimized for engagement. Individuation risks sliding into auto-repetition.

5. Proletarianization of temporal judgment

If Stiegler’s earlier concern was the proletarianization of industrial workers—their dispossession from the tools and sequences that once supported their know-how—the contemporary analog is the proletarianization of temporal judgment (Stiegler, 2010). Judgment here means more than “deciding what to watch.” It means pacing one’s attention, sequencing one’s tasks, plotting one’s projects, and composing a narrative arc out of discrete moments.

When platforms supply continuous micro-protentions (“Up next,” “You might like,” “People also bought”) and smooth over every transition with auto-play and nudges, they relieve individuals of the burden of deciding what comes next. In the short term, this feels like friction reduction. In the long term, it risks eroding the practical skills by which people once organized their own time: setting limits, enduring boredom, prioritizing, returning deliberately to what matters.

Again, this is not a moralistic defense of pre-digital asceticism. It is a structural point: if judgment about temporal sequencing is increasingly embedded in technical systems designed around engagement metrics, then the composition of biographical time—what you come to feel your life consists of—becomes partially programmable from the outside.

6. The pharmacological pivot

From a Gestalt Logos perspective, it is crucial not to stop at critique. Stiegler’s insistence on technics as pharmakon remains a live resource (Stiegler, 1998). The same infrastructures that compress, fragment, or over-stimulate our temporal experience can also be repurposed as supports for study, collective reflection, and care.

• A feed can be tuned (or hacked) to foreground long-form content, slow series, and revisitable archives rather than only the newest, hottest, or shortest items.

• Notification systems can be redesigned to batch and delay, encouraging chunked time for focused activity instead of continuous interruption.

• Recommendation engines can be constrained by user-defined horizons (e.g., “Help me stay with a single topic for a month”) rather than open-ended engagement.

These are not utopian fixes; they are pharmacological gestures—small shifts in tertiary retention that aim to restore some measure of temporal authorship to individuals and communities. In Stiegler’s terms, they contribute to new “associated hypomnesic milieus”: environments in which exteriorized memories and predictions are co-configured by their users instead of being imposed on them.

Part II, then, reframes platforms and AI systems not as generic “technologies” but as time-factories: industrial complexes that fabricate the forms through which time is lived, remembered, and anticipated. In Part III, Gestalt Logos will take up the more constructive task: sketching criteria and design heuristics for technical forms that do not merely capture attention, but cultivate temporal intelligence—the shared capacity to inhabit, narrate, and reshape the time we are given.

IV. GESTALT LOGOS HEURISTICS FOR TEMPORAL INTELLIGENCE

If platforms are time-factories, then the political and philosophical question becomes: How do we design for temporal intelligence rather than mere retention? Stiegler’s pharmacological stance refuses the fantasy of a “return” to a pre-technical innocence; the only way out is through technics—through new arrangements of tertiary retention that can sustain individuation, care, and inheritance (Stiegler, 1998, 2010). Gestalt Logos names one attempt to articulate criteria for such arrangements.

What follows is not a blueprint but a set of heuristics: compositional guidelines for evaluating and (re)designing technical systems at the level of time. They are meant to be applied both to personal practice (how we configure our own environments) and to institutional / platform design (how infrastructures are built).

1. From “more attention” to better-shaped time

The first move is definitional. As long as attention is framed as a scarce resource to be maximized, any design question will tilt toward “How do we get more of it?” Gestalt Logos suggests inverting the metric: the goal is not more attention but better-shaped time.

“Better-shaped” here does not mean more efficient or more productive in the economic sense. It means temporal structures that:

• allow for depth (staying with a topic, relationship, or practice beyond algorithmic novelty cycles),

• preserve rhythms (recognizable patterns of work, rest, gathering, solitude), and

• support inheritance (the ability to receive, modify, and pass on shared forms—stories, skills, commitments).

An app, platform, or practice is pharmacologically positive to the extent that it expands these dimensions rather than compressing them into a smooth continuum of micro-updates (Citton, 2017; Stiegler, 2010).

Heuristic 1: When evaluating a technical system, ask not “How engaging is it?” but “What kinds of temporal forms does it make likely or unlikely?”

If the default usage pattern is a jittery oscillation between many micro-episodes, the system is probably aligned with industrial retention rather than temporal intelligence.

2. Make time visible again

Industrialized tertiary retention tends to hide its own cuts. Infinite scroll, autoplay, and frictionless switching conceal transitions that might otherwise serve as moments of reflection. A core Gestalt Logos principle is therefore: make time visible.

Concretely, this can mean:

• Displaying session histories in ways that foreground continuity (“You’ve spent 40 minutes on this topic across three sessions”) rather than only raw totals (“4 hours screen time”).

• Introducing articulated intervals—clear beginnings and endings to episodes, with optional prompts to name what just happened (“What did you actually do in this block?”).

• Allowing users to annotate their own time within a system, turning usage into a narrative rather than a blur.

The point is not to shame users into less use but to re-insert phenomenological edges into otherwise continuous flows. Husserl’s analysis reminds us that experience is always structured as a synthesis across “nows” (Husserl, 1991). When platforms erase the felt difference between “this now” and “the next now,” they quietly appropriate that synthesis for themselves.

Heuristic 2: Good design exposes the user to the form of their time, not just the content in it. It supplies handles for narration (“this session,” “this phase,” “this series”) instead of dissolving everything into one long scroll.

3. Bias toward revisitable over disposable traces

Not all tertiary retentions are equal. Some are meant to be revisited—books, long-form essays, recorded lectures, scores. Others are designed for near-instant obsolescence—stories that expire, snaps that auto-delete, posts that sink beyond retrieval. The latter can certainly have value, but when a whole media ecosystem is tilted toward disposability, secondary retention (remembering and reactivating) loses its scaffolding.

A Gestalt Logos criterion is therefore to bias design toward revisitable traces:

• Make it easy to return to things: stable URLs, collections, series views, and archives that surface “what you’ve been building with” rather than only “what’s new.”

• Treat deletion and ephemerality as intentional choices, not defaults—tools for specific rituals (mourning, letting go, privacy), not generalized engines of forgetfulness.

• Reward re-reading and re-watching (e.g., by designing interfaces that recognize returns and support deeper layers of engagement) instead of only first clicks.

Stiegler’s worry about the industrialization of memory is not that we remember too much, but that we remember in formats we cannot appropriate—that our exteriorized memories are locked into logics of rotation and replacement (Stiegler, 1998). Revisitable traces are a technical precondition for Bildung: the slow shaping of self through repeated contact with forms.

Heuristic 3: Ask of any system: “What is it helping people return to?” If the answer is “mostly nothing,” its pharmacological tilt is likely toward disindividuation.

4. Restore user-configured protention

In Part II, we treated recommendation systems as engines for pre-filling protention: they decide what comes next before we do. Gestalt Logos does not reject recommendation as such; it asks who sets the horizons within which recommendation operates.

One way to restore temporal authorship is to provide user-configured protentional frames—parameters that define the kind of futures a system is allowed to propose. Examples:

• Topic horizons: “For the next 30 days, keep me within this theme (e.g., Hebrew, housing policy, violin technique). Don’t chase every new micro-trend.”

• Rhythm horizons: “Offer me content in 25-minute chunks with clear breaks,” or “Batch notifications into three windows per day.”

• Depth horizons: “Prioritize series and threads over singletons; help me stick with ongoing lines of inquiry.”

Instead of the system silently inferring and re-writing the user’s “goals” from behavioral data, protention becomes co-authored: a negotiation between human and machine parameters.

Heuristic 4: A system supports temporal intelligence to the extent that users can shape the space of likely next steps—not merely choose among options already optimized for engagement.

5. Embed collective time, not just individual time

Citton’s “ecology of attention” reminds us that attention is not purely private; it is structured by norms, institutions, and shared schedules (Citton, 2017). Likewise, temporal intelligence is not just a personal trick; it is a collective achievement. Calendars, liturgies, academic schedules, Shabbat, and civic holidays all function as macro-tertiary retentions: they give contour to the time of a group.

Gestalt Logos therefore looks for designs that link individual time to shared temporal forms:

• Tools that make it easy to align personal projects with collective rhythms (study groups, seasons, liturgical cycles, community events).

• Interfaces that reveal when many people are “with” a text, a topic, or a practice at the same time, not just who is “online.”

• Features that encourage co-authored archives—shared notebooks, public reading lists, collaborative timelines.

The point is not to produce a new universal calendar, but to resist the fragmentation into purely individualized time-lines each governed by private feeds. Stiegler’s associated milieus are, by definition, plural and collective: groups who share not just content, but the temporal structures in which content is encountered (Stiegler, 2010).

Heuristic 5: Evaluate systems by asking: “Does this help people inhabit shared time, or does it only optimize private streams?”

6. Design for care, not just control

Finally, Gestalt Logos insists that any meaningful reconfiguration of technics must foreground care. Stiegler’s later work casts care not as sentimentality but as a practice of investing in the futures of others—human and non-human, present and absent generations (Stiegler, 2010). Temporal intelligence is inseparable from this: to care is to shape how future time will feel for someone.

In practical terms, this suggests:

• Slowing down key transitions (e.g., from one video to the next, from one conversation thread to another) when stakes are high—conflict, grief, life decisions—rather than applying the same frictionless logic everywhere.

• Building in reflective pauses at structurally important points: endings of courses, completion of long threads, anniversaries of past events.

• Giving users rights over their retentions: the ability to export, annotate, and redeploy their data for their own projects, not only as raw material for corporate models.

Care, in this frame, is not opposed to technics; it is a mode of technical configuration. A platform that treats users only as objects of prediction cannot care for them, because it never lets them appropriate their own time.

Heuristic 6: Ask whether a system’s temporal structures make it easier or harder to take care—of oneself, of others, of shared worlds. Where care is structurally impossible, critique is not only permitted but mandatory.

Coda: Gestalt Logos as Practice

These heuristics do not exhaust the problem; they sketch a way of seeing. Gestalt Logos is less a doctrine than a practice of reading technical systems as temporal compositions. To look at a new app, platform, or policy through this lens is to ask:

• What retentional flows is this arranging?

• What kinds of Gestalten—stories, habits, institutions—does it make easier to form?

• Which possibilities of pacing, remembering, and expecting does it open, and which does it quietly close?

Stiegler’s provocation is that these are not side questions; they are the question, whenever technics is at issue. Our wager is that by integrating his retentional theory with Gestalt attention to form, and by extending both into the contemporary landscape of platforms and AI, we can begin to articulate a politics of time adequate to the present.

The task is not simply to log off, nor to celebrate every new affordance. It is to participate—critically, imaginatively—in the ongoing design of the time-factories we already inhabit.

References

Citton, Y. (2017). The ecology of attention. Polity Press.

Husserl, E. (1991). On the phenomenology of the consciousness of internal time (1893–1917) (J. B. Brough, Trans.). Kluwer Academic.

Stiegler, B. (1998). Technics and time, 1: The fault of Epimetheus (R. Beardsworth & G. Collins, Trans.). Stanford University Press. (Original work published 1994)

Stiegler, B. (2010). Taking care of youth and the generations (S. Barker, Trans.). Stanford University Press.

Zuboff, S. (2019). The age of surveillance capitalism: The fight for a human future at the new frontier of power. Profile Books.