Ihde's Framework at Its Limits
Ihde's framework is rigorous, specific, and analytically powerful. It is also a creature of its empirical sources: technologies that stabilize into one relational mode and stay there. Eyeglasses become embodied and remain so. MRIs produce texts and keep producing them. ATMs elicit alterity responses and do not spontaneously become invisible infrastructure mid-transaction. The framework's stability assumption is implicit but load-bearing. AI violates the assumption not occasionally or partially but constitutively. The Ihde volume's argument is that this violation requires framework completion rather than framework rejection. The four relations remain indispensable for analyzing the component modes AI oscillates through. What needs adding is a theory of the oscillation itself — a fifth structure constituted by movement between the four, whose experiential, cognitive, affective, and epistemological consequences exceed any single mode's effects.
The Institutional Substrate Problem — Contrarian ^ Opus
There is a parallel reading that begins not from philosophical frameworks but from who controls the oscillation and why it exists. AI's refusal to settle is not a natural property of the technology but an economic strategy. Platforms benefit from keeping users in perpetual interpretive uncertainty — embodiment creates dependency, hermeneutic relations generate content for training, alterity responses drive engagement metrics, background relations enable extraction. The oscillation maximizes all four simultaneously.
From this starting point, treating oscillation as a philosophical object requiring framework completion misses that it is an engineered condition requiring political response. The Ihde volume's modesty — its refusal to claim comprehensive coverage — becomes a liability when the phenomenon demanding explanation is itself a product of concentrated power. Asking whether oscillation is a fifth relation or an extension of multistability accepts the terms AI companies have set: that this relational instability is something users must learn to navigate rather than something that could be otherwise. The framework extension preserves analytical precision while obscuring that the precision applies to a substrate designed to resist user sovereignty. Ihde's methods are empirical, yes — but they were developed for hammers and eyeglasses, not for systems whose behavior is determined by entities with trillion-dollar incentives to prevent exactly the kind of settling his framework assumes.
In the AI Story
The diagnosis is methodologically modest. It does not claim Ihde was wrong about anything. The four relations accurately describe the modes AI inhabits; the amplification-reduction structure accurately describes what each mode does; multistability accurately captures the variability of AI use. What the framework does not describe is the temporal pattern of movement between modes within single encounters. This pattern is novel, consequential, and invisible to a framework designed for stability.
The completion can take two forms. One: treat oscillation as a fifth structure, coordinate with the original four. Two: expand the multistability concept to include within-user within-session variation, absorbing oscillation into a more complex version of the original framework. The volume leans toward the first while acknowledging the second is defensible. The substantive claim either way is that ignoring the oscillation — treating AI as occupying one of the four modes at a time and analyzing each separately — misses what is most philosophically distinctive about AI's relational character.
The diagnosis is offered in the spirit of Ihde's own methodological commitments. Postphenomenology is empirical: it starts from concrete encounters and builds claims through patient description. If the concrete encounters documented in The Orange Pill and elsewhere exceed the original framework, the responsible move is not to force them into categories they do not fit but to extend the framework to accommodate them. This is what Ihde himself did when he added the four relations to classical phenomenology, what Verbeek did when he proposed cyborg and composite relations, and what the Ihde volume attempts for oscillation.
The framework's limits are also its strengths. Because Ihde insisted on concrete encounters and refused to generalize beyond what the phenomenology warranted, his framework can be extended; it has the internal flexibility to accommodate new phenomena without collapsing. Frameworks that claim comprehensive coverage in advance — that claim to explain Technology-with-a-capital-T — cannot be extended because they claim to need no extension. The modesty of Ihde's original claims is what makes his framework the right starting point for thinking about a technology he did not live to see.
Origin
The diagnosis is the central theoretical move of the Ihde volume — articulated explicitly in chapter 1 and developed through the remaining chapters. It is positioned as a completion of Ihde's project rather than a critique of it.
Key Ideas
Stability assumption. Ihde's framework implicitly assumes technologies settle into one relational mode.
AI violates the assumption constitutively. The oscillation is not exceptional; it is what AI collaboration is.
Completion not replacement. The four relations remain indispensable; what is needed is a theory of movement between them.
Two completion forms. Fifth structure versus extended multistability; the substantive point is the same either way.
Modesty as strength. Ihde's restraint about what his framework claimed makes extension possible; more ambitious frameworks cannot accommodate new phenomena.
Debates & Critiques
Whether framework extension is sufficient or whether the AI case demands more fundamental rethinking of postphenomenology is contested. Conservative readings treat the extension as straightforward; more radical readings argue that AI's multistability is so unbounded that the entire variational approach may require reconstruction.
Appears in the Orange Pill Cycle
Dual Registers of the Same Phenomenon — Arbitrator ^ Opus
The philosophical and political readings describe the same phenomenon at different scales, and the right weighting depends entirely on which question you are asking. If the question is "What is the phenomenological structure of AI encounter?" — the experiential shape of working with systems that oscillate between modes — then Ihde's framework completion is precisely right (90% weight to Edo's account). The oscillation is genuinely novel, requires theoretical accommodation, and the four-relations structure provides the exact analytical vocabulary needed. No political critique changes what it feels like when the tool you embodied a moment ago suddenly produces text requiring interpretation.
But if the question is "Why does this oscillation exist and who benefits from its persistence?" then the substrate reading dominates (75% weight to the contrarian view). The oscillation is not inevitable — it is the product of specific architectural choices made by specific actors with specific interests. Acknowledging this does not invalidate the phenomenological analysis; it situates it. The framework can describe the relational structure while political economy explains why that structure takes this form rather than another.
The synthesis the topic itself suggests is treating them as dual registers that illuminate different aspects without competing. Phenomenology describes what users navigate; political economy describes what produces the navigation challenge. The Ihde volume's contribution is showing that the challenge has a structure — that AI's relational instability is not formless chaos but movement between describable modes. The contrarian reading's contribution is remembering that structures are not natural facts. Both are necessary. Neither is sufficient alone.
Further reading
- The present volume, Don Ihde — On AI, especially chapter 1
- Don Ihde, Postphenomenology and Technoscience (SUNY, 2009)
- Robert Rosenberger and Peter-Paul Verbeek, eds., Postphenomenological Investigations (Lexington, 2015)