The Schools Compared - A Working Map

Part 5 has covered nine schools across forty years of argument. This lesson does the accounting: what each established, where they disagree, what each is good for, and what they left open.

It is a synthesis rather than a new argument, and its purpose is to leave you with something you can use.

One case, nine readings.

A hospital trust introduces a new electronic patient record system. Two years later, clinicians hate it, patient outcomes are unchanged, the cost has doubled, and the trust is expanding its use.

A functionalist asks what functions the system performs for the trust as a system, and finds the manifest one (better records) has failed while latent ones have not: it satisfies regulators, produces auditable evidence of modernisation, and secures funding. Merton would call the persistence a case of latent function outliving manifest failure.

A conflict theorist asks whose interests it serves. Managers gained visibility over clinical work; clinicians lost discretion. The dispute over "usability" is a dispute over control of the labour process, conducted in technical language.

A power analyst asks which face is operating. Nobody was forced. But the decision to procure was taken in a forum with no clinical representation (second face), and the framing of the alternative as "staying on paper" made opposition look like backwardness (third face).

A symbolic interactionist watches what people actually do: the workarounds, the free-text fields used for the notes that matter, the informal handover conversations that carry the real information, and the negotiated order that has grown up around the official one.

A Goffmanian notes that the screen has restructured the consultation — the doctor now performs to two audiences, patient and record, and the record wins.

A critical theorist sees instrumental reason: a system optimised for what can be counted, displacing a communicative relation with a technical one. Habermas would call it colonisation of the lifeworld, and would predict exactly the pathologies reported.

A Gramscian asks how "digital transformation" became common sense, which organic intellectuals carried it, and what concessions were made to secure clinical consent.

A feminist asks who does the additional documentation work that the system created, and finds it falls on nurses and administrators — overwhelmingly women — whose time is not costed in the business case.

A Foucauldian notes that the system produces new objects of knowledge (risk scores, care pathways, flagged patients), makes clinicians visible to inspection, and turns the patient into a case constituted by fields.

A Bourdieusian observes that the clinicians resisting are those whose capital is embodied and clinical, and that the system revalues the field's stakes towards those with managerial capital.

And a structuration analyst notes that the system is not a thing acting on people: it exists only in the practices of those using it, and it is being remade daily by every workaround.

Every one of those readings is true. None is complete. And a study that used only one would be worse than one that used four.

That is what Part 5 was for.

The map

The nine positions, on four dimensions.

School	Society is	Order comes from	Change comes from	Best at
Functionalism	a system of interdependent parts	shared values, socialisation	differentiation	institutional interdependence, unintended consequences
Conflict theory	an arena of opposed interests	constraint and regulated conflict	conflict between groups	inequality, institutions as settlements
Symbolic interactionism	patterned joint action	continuous mutual interpretation	reinterpretation in encounters	how structures are actually produced; meaning
Critical theory	a system of domination through reason and culture	ideology, culture, the satisfaction of needs	critique and communicative reason	culture as power; the limits of instrumental reason
Gramsci	a hegemonic order	consent won in civil society	counter-hegemony, war of position	how common sense is made and unmade
Feminist theory	a gender order structuring everything	naturalisation of a constructed order	struggle, and the exposure of construction	the categories themselves; social reproduction
Structuralism / Foucault	systems of difference; discursive formations	classification; disciplinary and biopolitical technique	shifts in discursive formations	classification, categories of person, expertise
Bourdieu	fields structured by capital distributions	habitus fitting field; misrecognition	hysteresis, field crisis	reproduction of inequality without coercion
Late modernity	an order dissolving its own foundations	routine, expert systems, individualised responsibility	reflexive self-confrontation	risk, individualisation, the present

The four permanent arguments

Part 5's disputes reduce to four, and each is unresolved.

1. Structure and agency

The positions: Structure determines (structuralism, Althusser). Agency produces (interactionism). Structure is inside the actor as disposition (Bourdieu). Structure is inside the act as instantiation (Giddens). Structure and agency must be separated analytically to study their interplay (Archer).

What is settled: that structures do not act. Any sentence in which "society" or "the system" does something must be unpackable into what people did (5.8.2).

What is not: whether structure can be studied independently of action, and how much of conduct is disposition, calculation, rule-following or interpretation.

The working position: Archer's analytical dualism is the most usable, Bourdieu's habitus the most explanatory of inequality, and Giddens's duality the best corrective against reification. Use the one that fits the question.

2. Consensus and conflict

What is settled: Dahrendorf's answer (5.2.1). Both faces are real; no single theory holds both without becoming empty; the researcher selects by question.

What is not: how much of compliance is belief, how much is calculation, and how much is the absence of alternatives. Gramsci's hegemony, Abercrombie's dominant ideology thesis and Scott's hidden transcripts (5.5.1) mark out a genuine empirical dispute that varies by case and has not been resolved in general.

3. Material and ideal

What is settled: that the question "which comes first" is badly posed, and that what is wanted is an account of how the two are related in specific cases — elective affinity, hegemony, relative autonomy, the sociology of knowledge.

What is not: how much autonomy culture has, and whether discourse constitutes material relations or is constrained by them. This is the live edge between Foucault and every materialist tradition, and it has not moved much since the 1980s.

4. Explanation and interpretation

What each school is actually good for

The toolkit, stated as guidance.

Reach for functionalism when analysing institutional interdependence, unintended consequences, or why an arrangement persists despite failing at its stated purpose. Not for explaining conflict or change.

Reach for conflict theory when interests are visibly opposed, when analysing institutions as settlements of past disputes, or when asking who benefits. Not for explaining cooperation or the absence of conflict.

Reach for the three faces of power whenever anyone claims power is or is not operating. Always ask which face and what evidence.

Reach for interactionism and Goffman when the official account of a setting differs from what happens; when studying how a category is applied to a person; when the question is about meaning, identity or the production of order in encounters. Not for explaining why parties arrive so unequally equipped.

Reach for critical theory when analysing culture industries, instrumental reason, audit and metrics, or the political effects of satisfaction. Not when you need an account of audience agency.

Reach for Gramsci when asking how something became common sense, why consent persists, or why a movement failed. Always name the institutions doing the work.

Reach for feminist theory when examining any category — because the question "whose experience was this built from?" applies to all of them — and always for social reproduction, care, and the unmeasured.

Reach for Foucault when asking where a category of person came from, how expertise operates, how records and examinations constitute their objects, or how power operates through freedom rather than against it. Not when you need to explain who benefits or what should be done.

Reach for Bourdieu when explaining reproduction of advantage without coercion, when analysing taste, credentials, or professional fields, or when the question is why people exclude themselves.

Reach for late modernity theory for risk, individualisation, expertise crises, intimacy, and the shifting of responsibility onto individuals. Check the class evidence before accepting detraditionalisation claims.

Question: why do so few working-class students apply to elite universities?

A bad answer uses one framework. "Because of habitus" is Bourdieu applied as a label rather than a tool. "Because of hegemony" is worse.

A good answer uses several, in order, and says which does what.

Start with the distribution — what actually happens. Application rates by background, controlling for attainment. This is descriptive and it constrains everything else.

Then the structural conditions — admissions criteria, information availability, geographic distance, cost, the composition of school advice. Second face of power (5.2.2): what is kept off the table before anyone chooses.

Then the dispositional account — Bourdieu on habitus and the practical sense of the possible, with Lareau's evidence on concerted cultivation as the mechanism (5.8.1).

Then the interactionist account — what actually happens in the conversations where this is decided: the teacher's framing, the parent's response, the peer group's reaction. This is where the disposition becomes a decision, and it is observable.

Then the institutional account — Foucault on how "potential" is assessed and recorded; how the category "not university material" is produced and applied.

And then the reflexive check — feminist and standpoint theory: whose categories are "elite," "aspiration" and "attainment"? Who built the measurement?

Six frameworks, each doing a different job, none doing all of them.

And the discipline this requires: at each step, say what evidence would show the account is wrong. That is what stops a multi-framework analysis from becoming a list of things one might say.

Go deeper

What Part 5 established

Five things that are now background assumptions, and it is worth naming them because they were contested and are no longer.

Societies contain groups with opposed interests, and institutions distribute advantage. Conflict theory won so completely it dissolved as a school (5.2.1).

Structures do not act. Every structural claim must be unpackable into what people do (5.8.2).

Categories are historical products with authors, purposes and effects. Foucault, feminism, structuralism and the sociology of knowledge converge here (5.6.1, 5.7.2).

Culture is a site of power, not a reflection of something more real. Gramsci, the Frankfurt School and cultural studies established this against both functionalism and orthodox Marxism (5.4.1, 5.5.1).

And meaning is required for explanation of human conduct. Weber's demand became the interpretive tradition's programme and is now assumed even by quantitative researchers who would not put it that way (5.3.1).

What Part 5 left open

The agency problem in its hardest form. If subjects are produced by discourse (Foucault), disposed by habitus (Bourdieu), and hailed by ideology (Althusser), what exactly is doing the resisting? No framework in this Part answers this satisfactorily, and Habermas's attempt (5.4.2) is contested precisely on these grounds.

The normative problem. Critical theory, feminism and Gramsci all make evaluative claims. Weber's value-freedom (4.3.9) says they cannot be grounded in social science. Foucault says the grounds are themselves effects of power. Habermas's answer — that the standard is implicit in communication — is the only systematic attempt, and it is not generally accepted.

The scope problem. Almost everything in Part 5 was produced in Western Europe and North America, about Western European and North American societies, by people trained in a handful of institutions. The frameworks are presented as general and their range has not been established. Part 6 is where this is confronted.

And the connection problem. Merton's middle-range programme won (5.1.2), which means the discipline has many good partial theories and no account of how they relate. Parsons's question — how do these findings fit together? — was dismissed rather than answered, and sixty years on it is still open.

And an honest note on Part 5's own selection

This Part covered nine schools. It could have covered fifteen.

Not treated, or treated only in passing: rational choice theory and analytical sociology (5.3's foil, and a substantial programme in its own right); network analysis as a theoretical tradition rather than a method; actor-network theory and science and technology studies (Latour, Callon, Law), which is among the most influential bodies of work of the last thirty years; systems theory beyond Luhmann; postcolonial theory, which Part 6 takes; queer theory beyond Butler; ecological and posthumanist theory; and the pragmatist revival (Boltanski and Thévenot's sociology of critique).

The selection was made on the basis of what a beginner most needs to read the rest of the discipline, and it reflects a judgement that can be argued with.

Which is itself the lesson of 4.6.4 and 5.5.1 applied to this course: a curriculum is a selection made by someone, for reasons, with consequences for what its reader will find obvious. Stating that is the minimum a course owes.

"You have to choose a theoretical framework." Working sociologists routinely use several, selected by question. Partisanship is for people with something to defend.

"The schools are competitors." They are mostly answers to different questions that have been arranged into a competition by textbooks. Where they genuinely conflict — on agency, on the status of discourse — the conflicts are identifiable and few.

"Later schools superseded earlier ones." Functionalism's unintended consequences, conflict theory's opposed interests, interactionism's negotiated order and Gramsci's hegemony are all in current use. The sequence is not a progression.

"Theory is separate from research." Merton's whole programme was that theory and research must constrain each other, and the best work in this Part — Bourdieu's, Goffman's, the Frankfurt School's authoritarian personality studies — is empirical.

"These frameworks are universal." They were produced in a handful of Western institutions about a handful of Western societies. Their range is an open question, and Part 6 opens it.

Why it matters

You now have nine frameworks and a rule for selecting between them: match the tool to the question, name the framework you are using, and state what would show it wrong.

You can read the discipline's arguments. When a paper claims something is functional, or hegemonic, or discursively produced, or a matter of habitus, you know what is being asserted, what evidence would support it, and what the standing objections are.

And you know what is not settled. Structure and agency; consensus and conflict; material and ideal; explanation and interpretation; the grounds of critique; the scope of the frameworks; and how the discipline's many partial theories relate. A course that left you thinking these were resolved would have taught you the opposite of the subject.

Nine schools, mapped on four dimensions — what society is, where order comes from, where change comes from, and what each is best at.

Four permanent arguments, each unresolved: structure and agency (settled: structures do not act; open: whether structure can be studied independently of action); consensus and conflict (settled: both faces are real and the researcher selects by question; open: how much of compliance is belief); material and ideal (settled: "which comes first" is badly posed; open: how much autonomy culture and discourse have); and explanation and interpretation (settled: Weber's double requirement; open: what to do when the two conflict).

The toolkit, as guidance: functionalism for interdependence and unintended consequences; conflict theory for opposed interests; the three faces whenever power is claimed; interactionism when the official account differs from practice; critical theory for instrumental reason and culture industries; Gramsci for how something became common sense; feminist theory for the categories themselves and for social reproduction; Foucault for where a category of person came from; Bourdieu for reproduction without coercion; late modernity for risk and individualisation.

What Part 5 established: opposed interests and institutional advantage; that structures do not act; that categories are historical products; that culture is a site of power; and that meaning is required for explanation.

What it left open: what does the resisting, if subjects are produced; what grounds critique, if values cannot be derived; whether frameworks built in a handful of Western institutions are general; and how the discipline's many middle-range theories relate to one another — Parsons's question, dismissed rather than answered.

And Part 5's own selection was a selection, made by someone, for reasons, omitting rational choice, network theory, actor-network theory, queer theory beyond Butler, and the pragmatist revival. Stating that is the minimum a course owes.

One — run the hospital exercise. Take an institution you know and write one paragraph on it from four of the nine positions. Note which produces something you did not already know.

Two — name your default. Identify the framework you reach for first, then analyse something using its sharpest opponent. The discomfort is the exercise.

Three — build your own selection rule. For five questions you actually care about, write down which framework you would use and why. That list is what Part 5 was for.

Four — find an unresolved argument in the wild. Take any current sociological dispute and locate it on one of the four permanent arguments. Most disputes are instances.

Five — audit this course. Identify a school you think should have been included, and say what its absence costs. If you can do that, you have read Part 5 properly.

Part 5 is complete. The schools built the frameworks the discipline uses — and every one of them was produced in a handful of institutions, in a handful of countries, about a handful of societies, and presented as general.

This course has said four times that a canon is an institutional product and that exclusion carries an epistemic cost. Part 6 is where that argument stops being a caveat and becomes the subject.

Part 6 — THE OTHER SOCIOLOGIES takes the traditions the discipline built without: Indian sociology and the analysis of caste; Chinese and East Asian social thought; African and Latin American traditions; Islamic social science; postcolonial and decolonial theory; and the question of whether a genuinely global sociology is possible or whether the term is a contradiction.

It opens with the argument that makes the rest necessary: 6.1 — Orientalism and the Production of the Other.