Actor-Network Theory (ANT)

Overview

Actor-Network Theory treats human and non-human entities symmetrically as "actants" that form networks through processes of translation. Developed by Latour, Callon, and Law, ANT traces how heterogeneous networks are assembled, stabilized, and sometimes dissolved — rejecting the a priori distinction between the social and the technical.

When to Use

• Mapping how a technology, innovation, or practice became accepted (or failed)
• Analyzing the role of artifacts, standards, or devices in stabilizing social arrangements
• Tracing controversy and network-building in science and technology
• Understanding why a seemingly good innovation failed to gain adoption

When NOT to Use

• When the analysis requires strong normative judgments (ANT is descriptive, not prescriptive)
• When macro-level structural explanations are needed (ANT resists pre-given social categories)
• When non-human agency is irrelevant to the research question

Assumptions

IRON LAW: Non-human actors have AGENCY in ANT — treating technology
as a passive tool violates the framework's core principle. If your
analysis strips agency from artifacts, you are NOT doing ANT.

Key assumptions:

1. Generalized symmetry — human and non-human actors are described in the same analytical terms
2. No a priori distinctions between the social, technical, and natural
3. Networks are the unit of analysis, not individuals or structures
4. Stability is an achievement, not a given — networks require continuous maintenance

Methodology

Step 1: Identify the Controversy or Innovation

Select the phenomenon to trace. Follow the actors — do not impose pre-existing categories.

Step 2: Map the Actants

List all relevant human and non-human actors (people, organizations, technologies, documents, standards, natural entities) involved in the network.

Step 3: Trace the Four Moments of Translation (Callon, 1986)

Moment	Description
Problematization	A focal actor defines the problem and positions itself as an obligatory passage point
Interessement	Devices and strategies lock other actors into proposed roles
Enrollment	Actors accept and perform their assigned roles in the network
Mobilization	Enrolled actors come to represent wider constituencies; the network stabilizes

Step 4: Assess Network Stability

Evaluate whether the network holds, noting points of resistance, betrayal, or dissolution.

Output Format

## ANT Analysis: [Context]

### Focal Actor and Problematization
- Focal actor: [who/what defines the problem]
- Obligatory passage point: [the framing that makes the focal actor indispensable]

### Actant Map
| Actant | Type | Role in Network | Interests |
|--------|------|-----------------|-----------|
| [name] | [human/non-human] | [role] | [what they want] |

### Translation Process
1. **Problematization**: [how the problem was defined]
2. **Interessement**: [devices used to lock actors in]
3. **Enrollment**: [how actors accepted roles]
4. **Mobilization**: [how representatives stood for wider groups]

### Network Stability Assessment
- Stabilizing factors: ...
- Points of fragility: ...
- Black boxes formed: ...

### Implications
1. [Key insight about the network]
2. [What would happen if key actants were removed]

Gotchas

• Do NOT treat non-humans as mere "context" — they must have equal analytical weight
• ANT does not explain WHY networks form; it describes HOW they form
• Avoid "network" as metaphor — in ANT, networks are traced empirically, not assumed
• The researcher must "follow the actors" rather than impose categories from above
• ANT has been criticized for lacking normative power — pair with critical theory if evaluation is needed
• Black-boxing occurs when a network becomes so stable its internal workings become invisible

References

• Callon, M. (1986). Some elements of a sociology of translation. In J. Law (Ed.), Power, Action and Belief (pp. 196-233). Routledge.
• Latour, B. (2005). Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford University Press.
• Law, J. (1992). Notes on the theory of the actor-network. Systems Practice, 5(4), 379-393.