Sociology vs Psychology

You now know what sociology looks at and the arguments it has with itself. A fair sceptic — a friend, a parent, an interviewer — will ask the obvious next question: isn't that just psychology? Or economics? Or history? Or common sense with longer words?

This topic answers, one neighbour at a time. The goal is not to defend a territory. It is to make you precise about what question you are asking, because most confused thinking in this area comes from people mixing up two perfectly good questions.

We start with the closest neighbour and the most frequent confusion.

Nisha is nineteen, in her second year of college, and she has stopped attending. She sleeps badly, cannot concentrate, has withdrawn from her friends, and describes a heaviness she cannot explain. Her family, who love her, do not know what to do.

Two researchers are asked to help understand what is happening.

The first talks to Nisha for an hour. She asks about her sleep, her appetite, her thoughts about herself, whether this has happened before, whether it runs in the family. She listens for patterns of thinking — the certainty that she is a burden, the anticipation of failure. She considers biology, temperament, the effect of a loss two years ago. She proposes a treatment aimed at Nisha: therapy, possibly medication, a structured routine.

The second never meets Nisha. He looks at data on the college: how many students report similar symptoms, and how that number has moved over five years. He finds that reported cases have doubled; that they cluster heavily among students in the first generation of their family to attend college; that they spike in the eight weeks before final examinations; that the college of four thousand students employs one counsellor; that student housing is overcrowded; and that the labour market these students are aiming at absorbs perhaps a fifth of them.

The first researcher is a psychologist. The second is a sociologist. Neither is doing the other's job badly. They are answering different questions.

Notice precisely what each question is.

The psychologist's question: why is this person suffering, and what will help her? The unit is one mind, one body, one history.

The sociologist's question: why has the rate doubled, and why is it distributed like that? The unit is a population and a set of arrangements.

Here is the crucial point. The second question cannot be answered by studying individuals one at a time, however carefully. If you interview two hundred students with depression, you will get two hundred detailed personal accounts — and not one of them will mention that the counsellor-to-student ratio is one to four thousand, because no individual experiences a ratio.

And the first question cannot be answered by studying rates. A ratio does not tell you what to say to Nisha on Tuesday.

Nisha needs both researchers. The confusion begins only when either claims the other is unnecessary.

The idea, in plain words

Durkheim drew the line on purpose

This boundary was not an accident of university administration. Durkheim (4.2) drew it deliberately, because sociology's existence depended on it.

His argument: if every social pattern can be explained by individual minds, sociology is a redundant branch of psychology and should be closed. So he had to show that some facts are irreducibly social. His demonstration is one of the most consequential in the history of the social sciences.

Take suicide — the case that looks most purely psychological. Durkheim conceded everything about the individual case: each suicide has its own despair, its own biography, its own proximate cause. But then he asked a question no psychology can answer. Why is the rate stable? Why does a society produce roughly the same number each year, and why do some categories of people reliably produce more than others, year after year, decade after decade?

A rate is not a property of any individual. It is a property of a group. And Durkheim's answer was also social: variation in how tightly people are integrated into groups, and how well their expectations are regulated. Hence his famous types — egoistic (too little integration), altruistic (too much), anomic (too little regulation), fatalistic (too much).

He gave the error a name: psychologism — explaining social facts by individual psychology alone. His rule was stark: the determining cause of a social fact must be sought among preceding social facts. Modern sociologists usually treat that as too strict. But the underlying point stands, and it is the reason the two disciplines are separate.

And then psychology returned the favour

Here is what makes this comparison genuinely interesting rather than territorial. Some of the strongest support for the sociological view of human beings came from psychology — from experiments showing that situations overwhelm dispositions far more than anyone expected.

Solomon Asch (1950s) had participants judge which of three lines matched a reference line — a trivially easy task. Seated among confederates who confidently gave a wrong answer, a substantial proportion of participants went along with the group, some of them repeatedly. The judgement of one's own eyes bent under social pressure.

Stanley Milgram (1960s) found that a majority of ordinary people, instructed by an experimenter in a lab coat, would administer what they believed were escalating electric shocks to a stranger. The finding was not that people are cruel; it was that behaviour tracked the structure of the situation — the authority's proximity, the victim's visibility, whether any other participant refused first.

Philip Zimbardo's 1971 prison simulation is the most famous of these studies and should be handled with care, because it has not aged well: later examination of the archives showed guards were coached, the sample was self-selected, and the design was closer to improvised theatre than experiment. Learn it as a cautionary tale about both the power of roles and the seductions of a vivid result. Milgram's studies have also been reassessed — variations in his protocol were not fully reported, and obedience rates were less uniform than the popular version suggests.

Even with those caveats, the accumulated body of situationist research supports something a sociologist recognises immediately: change the arrangement — who is watching, who is in charge, what the role licenses, who else has spoken — and ordinary people behave very differently.

Psychology has even named the mistake that this research corrects. The fundamental attribution error is our systematic tendency to explain others' behaviour by their character and underrate the situation. Which is, near enough, a psychological account of why the sociological imagination has to be learned (1.1.1). It does not come free.

Overlap: where the two disciplines actually meet

The boundary is real but not a wall, and the interesting work often happens on it.

Social psychology is a field that both disciplines claim, with two distinct traditions: a psychological social psychology (experimental, individual-focused) and a sociological social psychology descending from Mead and symbolic interactionism (2.3, 5.3), focused on how selves and meanings are produced in interaction.

The sociology of mental health (9.6) is a mature field built on exactly the distinction in our opening story. Its classic work — George Brown and Tirril Harris's Social Origins of Depression (1978) — showed that the onset of depression among women in London followed identifiable social patterns: severe life events, plus "vulnerability factors" such as lacking a confiding relationship, having several young children at home, and no employment outside it. Not instead of psychology. Underneath it.

The sociology of emotions (12.1) — Hochschild's emotional labour and feeling rules, Collins on emotional energy — treats feelings as socially patterned rather than purely inner events.

Cognitive sociology studies socially shared classification: how groups come to divide time, kinship, purity, colour and normality in particular ways.

Three hundred students sit an examination. Forty do very badly.

Psychological questions: How does anxiety impair working memory? Why does this student freeze and that one focus? What thinking patterns predict poor performance under pressure? What intervention helps an individual student?

Sociological questions: Are the forty randomly distributed, or concentrated by school background, language, gender, caste or income? Did the paper assume knowledge available only through paid coaching? How many hours of unpaid housework did each student have last week? Was the medium of instruction anyone's first language?

Now the important part. Suppose the forty turn out to be overwhelmingly first-generation students studying in a second language while doing four hours of domestic work a day. The psychological findings remain completely true — anxiety does impair working memory. And they would be the wrong basis for policy, because the intervention that follows from them is relaxation training, while the intervention that follows from the sociology is a change in the examination, the language, or the household expectation.

Same forty students. Same evidence. Different level, different lever.

Go deeper

Two directions of error

Psychologism — explaining a social pattern purely by individual minds. "Unemployment is high because people lack motivation." "The caste system persists because people are prejudiced." Both may contain something; neither explains a distribution, and both quietly turn a structural question into a question about the character of the people at the bottom.

Sociologism — the mirror error: explaining everything by social structure and leaving no room for individual mind, variation or responsibility. It has its own lesson later in this Part (1.5.3), because it is the vice a new sociology student is most at risk of.

The tell for both is the same: a single-level explanation offered with great confidence for a phenomenon that visibly operates on two levels.

Why "it's psychological" often functions politically

Worth noticing, because you will see it constantly. When a patterned problem is redescribed as a psychological one, the implied remedy moves from the arrangement to the person.

If burnout among nurses is a matter of individual resilience, the response is a mindfulness programme. If it is a matter of staffing ratios, shift patterns and unpaid overtime, the response costs money. Both descriptions can be evidenced. Only one requires the institution to change.

This is not an argument against psychology, which has been perfectly clear for decades that chronic workplace stressors are structural. It is an argument about who gets to choose the frame — and it is 1.1.1's trouble/issue distinction, arriving in an institutional setting.

The honest limits of the sociological view

Sociology can be one-eyed too, and a good student should be able to say how.

Sociology is weak at explaining variation within a position. Two children in the same family, same class, same school, same caste, same city — and one thrives while the other does not. Structure gives you the distribution; it does not give you the individual case, and pretending otherwise is overreach. Temperament, cognition, illness, chance and biography are real.

Sociology is also weak at mechanisms inside the head. If cultural capital produces confidence in a classroom, what actually happens in the child's attention, memory and self-concept? That is a psychological question, and answering it makes the sociology better rather than dispensable.

The mature position is not a truce but a division of labour: sociology explains the distribution and the arrangement; psychology explains the process inside the person; and the best work on any real problem — mental health, education, crime, addiction, discrimination — uses both.

"Sociology denies that individuals have minds." It denies that minds alone explain rates. Nothing in this course requires you to believe that Nisha's suffering is not really hers.

"Psychology is about individuals, so it ignores society." Not true of the field. Social psychology, developmental psychology and community psychology are all deeply concerned with social context. The disciplinary difference is about the primary unit of explanation, not about whether the other exists.

"The sociological explanation is the deeper one." Depends entirely on the question. For "what will help Nisha next week?" the psychological account is deeper. Depth is relative to the question asked, not a property of a discipline.

"Social factors mean nobody is responsible." 1.3.1 again: explanation and responsibility are different questions.

"Milgram and Zimbardo proved that anyone will do anything." Careful — both studies are weaker than their fame, and both have been substantially reassessed. Use them as illustrations, not as proofs, and say so.

Why it matters

It stops you from asking one question and answering another. Half the bad arguments about crime, poverty, education and mental health are two people answering different questions with equal confidence and increasing volume.

It makes you useful. In any real institution — a school, a hospital, an office — the person who can say "that is a distribution problem, not a motivation problem, and here is why" is doing something valuable that almost nobody else in the room is doing.

And it keeps sociology honest. A discipline that cannot say what it is not good at is not a discipline; it is an ideology. Knowing where psychology is stronger is part of being a competent sociologist.

Psychology and sociology study the same beings with different units of analysis: the individual mind versus groups, institutions and structures.

Durkheim drew the line deliberately with the study of suicide: a rate is a property of a group and cannot be explained by individual despair. He named the error psychologism.

Psychology returned the favour: Asch, Milgram and the situationist tradition show behaviour tracking the situation more than the disposition — though Zimbardo's prison study and parts of Milgram's work have been seriously reassessed and should be taught with those caveats.

The fundamental attribution error is psychology's own account of why the sociological imagination must be learned.

The disciplines overlap productively in social psychology, the sociology of mental health (Brown and Harris), the sociology of emotions and cognitive sociology.

Sociology's own limits: it explains distributions, not individual cases, and not the processes inside a person's head.

Unit of analysis — the entity a study is actually about: a person, a household, an organisation, a rate, a society.

Psychologism — explaining social facts purely in terms of individual psychology.

Sociologism — explaining everything in terms of social structure, leaving no room for individual variation or agency (see 1.5.3).

Social fact — Durkheim's external, general, coercive pattern that must be explained socially.

Situationism — the finding that behaviour is shaped more by features of the situation than by stable personal dispositions.

Fundamental attribution error — the tendency to over-explain others' behaviour by character and under-explain it by circumstance.

Social psychology — the study of how people think, feel and act in social contexts; pursued in both a psychological and a sociological tradition.

Emotional labour — Hochschild's term for the work of managing one's feelings to meet the requirements of a job.

Feeling rules — socially shared expectations about what one ought to feel in a given situation.

One — two lists. Take a difficulty you know well (in yourself or someone close). Write the psychological questions you would ask, then the sociological ones. Notice that the second list needs information about other people you may not have.

Two — catch the attribution error. For two days, note each time you explain someone's behaviour by what they are like. Then generate a situational explanation for the same behaviour. This is the sociological imagination as a daily exercise.

Three — spot the frame shift. Find one institutional problem currently being addressed with a programme aimed at individuals — resilience training, motivation workshops, awareness sessions. Ask what the structural version of the diagnosis would look like, and what it would cost.

Four — the sibling question. Think of two people from near-identical positions whose lives diverged. Write down what sociology can and cannot explain about the difference. Being able to say where your tools stop is a mark of competence, not weakness.