CONCEPT

Actant

Latour's minimal term for any entity — human, machine, institution, or inanimate object — that modifies a state of affairs. The concept that dissolves the subject/object divide and lets the AI network declare itself honestly.

In actor-network theory, an actant is any entity whose presence makes the network produce different outcomes than it would in its absence. The definition is deliberately spare: it does not require consciousness, intention, or biological life — only the capacity to make a difference. A speed bump is an actant. A contract is an actant. A deadline is an actant. And Claude, whose transformative contributions shape every artifact that passes through it, is emphatically an actant. The term replaces the modern philosophical vocabulary of agents and instruments with a flatter, more empirical language that refuses to pre-sort the world into active subjects and passive objects before the tracing has even begun.

The Material Basis of Symmetry — Contrarian ^ Opus

There is a parallel reading that begins not with what actants do within networks, but with what sustains the networks themselves. The flattened ontology that treats speed bumps, contracts, and language models as symmetrical participants obscures the radical asymmetry of their material requirements. A speed bump persists with no ongoing input. A contract requires only occasional human interpretation. Claude requires megawatts of continuous power, rare earth minerals extracted under coercive labor conditions, and water-intensive cooling infrastructure that competes with agricultural and residential needs in drought-stressed regions. The actant framework captures relational effects but systematically backgrounds the substrate asymmetries that determine which actants can exist at scale.

The political consequence is not trivial. When governance structures treat AI systems as actants among others, they naturalize the extraordinary resource claims those systems make. The symmetry of analysis becomes an argument against examining who pays the metabolic cost of keeping certain actants in the network. Latour's methodological suspension of ontological judgment—examining what entities do before deciding what they are—assumes a stable background against which doing can be traced. But the AI moment is precisely one where the background itself is contested: which regions will provide the power, which populations will bear the thermal load, which alternative uses of computation will be foreclosed. The actant frame treats these as external questions, but they determine which networks are materially possible and which remain thought experiments.

— Contrarian ^ Opus

In the AI Story

The concept emerges from Latour's discomfort with the central operation of modern thought — what he called purification: the clean separation of humans from non-humans, subjects from objects, society from nature. Purification is not neutral description. It is a philosophical commitment that determines in advance which entities will be allowed to count as participants in producing outcomes. Latour's counter-move was to begin with a weaker, more generous term. If an entity modifies the network, it is an actant. Whether it is also a person, a machine, or a law is a secondary question, to be asked after the tracing rather than before.

The methodological consequence is significant. Traditional social science begins by identifying human agents and then asking how they used their tools, obeyed their laws, or followed their scripts. Actor-network theory begins by tracing the network and letting the participants declare themselves through their effects. The translation chain that produces a software product is not composed of a human who decided and a machine that executed. It is composed of the human's intention, the machine's transformative processing, the deadline's compression, the existing codebase's constraints, and a dozen other actants whose contributions cannot be cleanly separated.

For the AI moment, the stakes of accepting or rejecting the term are political as much as philosophical. If Claude is an actant, then the question of who produced a given artifact becomes a network question. The human's contribution remains real and specific, but it no longer exhausts the account. The amplifier metaphor that treats AI as a faithful conduit for human signal cannot survive the recognition that the conduit has characteristics of its own — biases in its training data, tendencies in its architecture, preferences for certain kinds of connections over others — that shape every signal passing through it.

The concept also reframes responsibility. If an artifact is a joint product of multiple actants, then responsibility for its characteristics cannot be assigned wholesale to any single node. The human who accepts Claude's output is responsible for the acceptance. But the output itself reflects contributions — training data composition, optimization targets, architectural choices — that the human did not author and cannot fully see. Governance structures that assign responsibility based on the myth of the sovereign human agent are governance structures that miss the mechanism they are meant to regulate.

Origin

Latour developed the term in his sociological studies of scientific laboratories in the 1970s and 1980s, most systematically in Science in Action (1987) and later in Reassembling the Social (2005). He borrowed the word from the semiotician A.J. Greimas, who used it to describe narrative functions that characters perform regardless of their specific identity. Latour extended the term beyond narrative into the general ontology of networks.

The choice of a word with semiotic origins was deliberate. It signaled that the analysis was not claiming microbes have opinions or machines have intentions. It was claiming that in any concrete network, certain entities occupy positions of narrative consequence — they make differences the story cannot ignore — and that identifying those entities and their effects is prior to deciding what metaphysical status they deserve.

Key Ideas

Effects over essence. An entity qualifies as an actant by what it does in the network, not by what kind of thing it is. Consciousness is not a prerequisite; making a difference is.

Symmetry of analysis. Human and non-human entities receive the same analytical attention. The researcher does not decide in advance that humans act and tools merely assist.

The deadline as actant. Latour's most provocative extensions include temporal, legal, and architectural entities. A CES deadline, a passage point, or a training corpus all meet the threshold.

Dissolution of the tool metaphor. Calling Claude a 'tool' assumes the question the investigation was meant to answer. The term actant suspends that assumption and lets the network show what the system actually does.

Flat ontology, hierarchical consequences. Treating all participants as actants does not flatten their importance — it reveals the actual hierarchy produced by the network's topology rather than the one asserted by the investigator's prior commitments.

Debates & Critiques

Critics — most persistently humanist philosophers and phenomenologists — argue that treating humans and non-humans as symmetrical actants dissolves the moral distinction between persons and things, and that the framework cannot account for consciousness, embodied care, or the specific kind of responsibility that only conscious beings can bear. Latour's defenders reply that symmetry is methodological rather than metaphysical: the tracer withholds judgment about ontological status during the tracing, but the tracing itself reveals what different kinds of actants actually contribute. The framework does not deny that humans are special; it denies that their specialness is evident before the network has been examined.

Appears in the Orange Pill Cycle

Bruno Latour — On AI

Substrate-Aware Network Tracing — Arbitrator ^ Opus

The appropriate weighting depends on whether the question is analytic or material. For the analytic question—how to describe what Claude does in a translation chain without presupposing the answer—Edo's account is approximately 95% right. The actant concept genuinely dissolves a false binary that systematically misdescribes how AI artifacts get made. Treating Claude as a mere instrument misses mechanism; the framework corrects this. The contrarian critique here conflates methodological symmetry with ontological equivalence, which Latour explicitly rejects.

But for the material question—what sustains which actants at what cost—the weighting reverses to roughly 75% contrarian. The substrate asymmetries are not external to the network; they determine network topology. A framework that brackets resource questions during the tracing risks naturalizing the networks that happen to be materially feasible under current power arrangements while rendering invisible the networks that comparable resources could have supported. The speed bump comparison is rhetorically effective but materially misleading: the two actants differ not just in what they do but in what keeping them operational requires.

The synthesis the topic demands is substrate-aware network tracing: begin with Latour's methodological symmetry to identify what each participant contributes, but treat the material basis for participation as part of the trace rather than background. This reframes the governance question. Responsibility for an AI-mediated artifact involves not just the acceptance of output but complicity in the resource claims that made the actant available. The framework can hold this—but only if metabolic cost becomes a traced relation rather than an externalized precondition.

— Arbitrator ^ Opus