Who makes the categories by which we live?

A label that opens a door

Imagine a training center that divides its adult learners into “independent” and “dependent.” Learners earn the first label by completing workshop exercises without asking a tutor for help. Independent learners may use the workshop in the evening; everyone else must attend supervised sessions.

Asha notices a loose fitting on a machine and asks a tutor to check it. The request counts as help, so her file records it as dependence, and she loses evening access.

The center says it has simply recorded a fact. Its record also changes what Asha can do.

Michel Foucault studied how such practices make people into subjects: people classified and directed by others, who may also come to understand themselves through those descriptions. [1] Gadamer asked what inherited expectations bring to interpretation; Foucault asks how a category acquires the authority to organize someone’s life.

Power can make things happen

A tutor’s command to stop a machine is an obvious exercise of authority. Foucault also directs attention to arrangements that shape the actions open to people. He distinguishes these relations of power from physical capacities and from the communication of information, while showing how the three work together. [1]

Having lost her access, Asha may now hide her doubts to earn it back. Tutors count requests, and learners judge themselves by the resulting scores. No one needs to order Asha to become the sort of person who never asks for help; the arrangement makes that conduct attractive.

This is productive power: power that helps form conduct, abilities and identities. “Productive” does not mean beneficial. Training may give Asha real skills while its assessment encourages her to hide a sensible question, and commands alone cannot explain that combination. [2]

How a file becomes knowledge

The center can report precisely how many times Asha asked for help, and it then treats that count as evidence of a general dependence. A question about a loose fitting, a request to repeat an unclear instruction and an inability to perform a task all become the same kind of mark.

Power–knowledge names the connection between ways of knowing people and the practices that direct them. Assessment creates records; records support categories; categories authorize further assessment and intervention. Foucault’s account of examination brings these activities together. [2]

At the center, the score does more than describe a trait that exists independently of it. The rules establish what will count as dependence and then distribute opportunities by it. Those denied evening access get fewer chances to practice, and their later performance may reinforce the original judgment.

None of this makes every recorded fact false: Asha did ask for help. What remains questionable is the inference from that fact to a judgment about her character, with consequences for her access. Foucault explicitly distinguishes challenging the power of knowledge from refusing every verified truth. [1]

Asking how the category came about

Genealogical inquiry investigates how a practice became established through events that could have unfolded differently, which is what “contingent” means here. It asks about struggles, decisions and unintended effects, rather than assuming that today’s arrangement was bound to emerge. [2]

For our center, we could examine old assessment forms, funding rules, tutors’ accounts and learners’ objections. Did the score begin as a way to allocate scarce staff? When did it become a description of character? Were questions about safety once counted differently?

These are research questions, not a history we already know. And finding a practical origin would not by itself discredit the category: we would still have to ask whether it distinguishes what matters, and what its use does to learners.

Immanent critique could test the center against its promise to develop capable learners, while a genealogy would investigate how this particular measure of capability became authoritative. The two inquiries can assist each other without becoming one method, and neither requires imagining a single hidden planner who designed every consequence.

A different question about meaning

Jacques Derrida offers a limited contrast. In “Différance,” he argues that a concept gains its meaning through its relations to other concepts, rather than containing a complete meaning within itself. Différance joins differing with deferring: a sign has meaning through differences and references beyond what is immediately present. The changed letter is visible but inaudible in French. [3]

Consider “independent” in our example. It gets its point from contrasts with help, dependence and supervision. Yet even a capable learner relies on teaching, shared equipment and other people’s maintenance. That troubles an ideal of total self-sufficiency without erasing the useful distinction between performing a task with help and performing it alone.

This suggests one kind of deconstructive question: does a preferred term depend on what it places beneath itself? Deconstruction examines such tensions within a text or a conceptual opposition. Deconstruction is not destruction. Nor does it merely reverse the ranking and declare dependence always better. [5] Our small example does not establish Derrida’s whole argument.

Structuralism seeks explanatory relations within a system, rather than treating its elements in isolation. The later label poststructuralism groups varied challenges to the stability and foundations of such systems; it names no single doctrine. The two names are not interchangeable, and neither makes Foucault and Derrida one voice. [4] Here Foucault guides an inquiry into practices and their history, while Derrida questions the demand for a meaning that could stand entirely on its own.

A classification can be challenged

Wittgenstein’s language-games showed why a word’s role depends on what people do with it. At the center, understanding “independent” includes noticing who records the label and which doors it opens. This is a connection between lessons, not proof that Wittgenstein and Foucault share a theory.

Asha and other learners might propose assessing safety questions differently, or deciding access through a practical demonstration, and they would need evidence about skills and risks. Foucault’s account leaves room for resistance because those who are directed remain capable of responding. It does not promise that resistance will succeed. [1]

Changing a word alone may leave the assessment intact. Changing the assessment may alter both what staff know and what learners can become.

Try questioning the file

A manager says, “Nobody ordered Asha to hide questions, so power is not involved.” A learner replies, “Then every assessment must be false.” Explain both mistakes. Identify one thing a genealogy would need to investigate. This exercise is optional.

Read a possible response

The manager assumes that power works only through commands. The access rule encourages Asha to conceal questions by rewarding a particular kind of performance, and it turns a count into an identity that tutors and learners may come to accept; neither effect requires an explicit order. The learner makes the opposite mistake, since asking whether the classification encourages competent work is different from denying that a request was recorded accurately.

A genealogy could compare earlier forms and interview participants to find out why requests became the measure of dependence. It would have to establish that history, not invent a bad motive. Learners could then argue about the classification’s evidence and consequences. A category’s historical origin does not make it useless, and an accurate number or a good intention does not settle whether its use is justified.

Looking back: what a method brings into view

How does the way we ask a question shape what we can learn? Each of the module’s thinkers brings a different part of a problem into view. Peirce and James examined meaning, inquiry and truth through consequences. Frege separated sense from reference, and Wittgenstein examined the variety of language’s uses. Husserl and Merleau-Ponty investigated experience, and Beauvoir asked about freedom within unequal situations.

Horkheimer and Adorno questioned reason’s service to domination, and Gadamer examined the expectations we bring to understanding. Foucault and Derrida challenged categories and meaning in distinct ways. The inquiries overlap, but their aims and answers can conflict.

Two concepts from the module can be brought to bear on the training center. From Peirce comes the pragmatic maxim, which clarifies a claim through the differences it could make in practice: what should we expect if it is true? [6] [7] From Foucault comes productive power, which directs attention to how an arrangement shapes conduct and identities, including the conduct it then records as evidence. [1] [2] One asks what a label promises; the other asks what happens when people must live under it. Bringing both to one case does not reconcile the two philosophies.

Checkpoint: when fewer questions count as progress

Return to Asha’s training center. A manager says its assessment produces independent learners who can work safely. Recorded requests for help have fallen, but some learners tell Asha that they have left loose fittings unreported to protect their evening access. The manager replies, “The figures prove the assessment works.”

Use the pragmatic maxim and productive power to examine the claim, and describe a check the center could make. Then say what question these concepts leave open, and what the sources used here cannot show. This checkpoint is optional.

Read a checkpoint response

Pragmatic clarification asks what “independent learners who can work safely” should lead us to expect. Never asking for help is different from recognizing when help is needed, so a supervised demonstration could show whether learners recognize a fault and seek appropriate assistance. Fewer recorded requests cannot, on their own, distinguish growing competence from concealed uncertainty. Clarifying the claim tells us what needs investigating; it does not establish that the assessment succeeds.

Productive power helps explain why the figures can change without the promised improvement. The access rule makes silence a way to keep an opportunity, and so helps produce the very conduct the center then treats as evidence of independence. The count may be accurate even while the judgment based on it misleads. A new label would not by itself remove the incentive, whereas a different assessment could change both what learners do and what tutors learn about them.

That leaves a real question open: how much supervision protects learners without unnecessarily restricting their opportunities? The two inquiries expose a problem without deciding which arrangement would work best. And remember what kind of evidence we have been reading: the Foucault reading comes from a web transcription that was not checked against the printed text or another translation, and the selected philosophical passages supply no evidence about any real training center. Judging a real center would need records, observation and learners’ accounts.

Carry this forward

Ask what a category records, what it leaves out and what changes when people must live under it. The next module follows overlapping histories of global and diasporic philosophy. It opens with A self seen through a racial order, where Du Bois and Fanon examine racialized experience. These arguments were developing alongside the conversations in this module, not arriving after a finished European canon.

Sources and limits

[1] Michel Foucault, “The Subject and Power” (1982), for subjects, relations of power and resistance, in a transcription that identifies Hubert L. Dreyfus and Paul Rabinow’s Michel Foucault: Beyond Structuralism and Hermeneutics, University of Chicago Press, 1983, pages 208–226. Read the opening of “Why Study Power? The Question of the Subject” through the paragraph defining the two senses of “subject”; and the introduction and complete numbered subsections 1–2 of “How Is Power Exercised?”

The site identifies the first part as written in English and credits Leslie Sawyer with translating the second from French. These headings and paragraph boundaries locate the HTML reading; printed pages and translation were not independently collated.

[2] Gary Gutting and Johanna Oksala, “Michel Foucault”, Stanford Encyclopedia of Philosophy, revised 21 April 2026, for genealogy, disciplinary power and examination. Read the complete main text of §§3.3–3.4; here § marks a numbered section. This scholarly account supplies context for Discipline and Punish, which was not read as a whole.

[3] Jacques Derrida, “Différance,” in Richard Kearney and Mara Rainwater, eds., The Continental Philosophy Reader, Routledge, 1996, for différance and meaning within relations of difference. Read the bounded passage on pages 441–450, stopping at the paragraph ending “difficult and uncomfortable”; and notes 1–2 on page 464. The acknowledgments credit David B. Allison’s translation in Speech and Phenomena, and Other Essays on Husserl’s Theory of Signs, Northwestern University Press, 1973. Pages 441–446 introduce the spelling and senses of différance; 448–450 examine concepts within relations of difference. These are anthology page numbers. No separate edition or French original was compared.

[4] Kearney and Rainwater, editorial introductions and bibliographies in the same anthology, pages 336–338 and 438–440, read completely, for context on the two projects, including their caution against treating Foucault simply as a structuralist. The works listed were not separately read, and the introductions’ broad historical claims are not all adopted here.

[5] Leonard Lawlor, “Jacques Derrida”, Stanford Encyclopedia of Philosophy, revised 27 August 2021, for deconstruction. Read the first two main-text paragraphs of §5, through its account of the early two-phase approach and the names used to rethink an opposition. Later definitions and cited primary works were not separately read for this lesson.

[6] Charles Sanders Peirce, “How to Make Our Ideas Clear”, Popular Science Monthly 12, January 1878, pages 286–302, in the linked English transcription, for the pragmatic maxim in the recap and checkpoint. For the checkpoint, section II was reread completely, with the final paragraph of section IV, which distinguishes clarity from truth. The lesson on pragmatic clarification gives the earlier, wider reading scope. This is Peirce’s English; printed pages and diagrams were not compared, and his later revisions are outside this selection.

[7] Catherine Legg and Christopher Hookway, “Pragmatism”, Stanford Encyclopedia of Philosophy, substantive revision 30 September 2024, for context on pragmatic clarification and the differences within the tradition. The checkpoint revisits the complete main text of §1.1 and the opening of §2, before §2.1. Other works discussed there were not thereby read.

Sources consulted 18 September 2026, and the checkpoint sources rechecked 19 September 2026. Selected anthology page images were checked. Asha, the center, its labels and possible reforms are original teaching material, and the checkpoint’s falling request count and unreported fittings extend that invented case. The example establishes no actual institution’s history, and the checkpoint’s comparison does not establish a shared theory. The reading does not cover all of Foucault or Derrida, every structuralist approach, or a complete historical case study. No independent scholarly review or learner testing is claimed.