Sociology as a Science
Sociology as a Science: The Debate
Can sociology be a science? Should it try to be? This is one of the most fundamental debates in the discipline and appears regularly on the AQA Theory and Methods paper.
What is Science?
Before asking whether sociology can be scientific, we need to define what science is. There are several competing definitions.
The Conventional View: Positivism and the Hypothetico-Deductive Method
The conventional view of science involves the hypothetico-deductive method:
1. Observation: the scientist observes a phenomenon
2. Hypothesis: they develop a testable prediction to explain it
3. Experiment: they design an experiment to test the hypothesis, controlling variables
4. Data collection: quantitative data is gathered
5. Analysis: the data is analysed to determine whether the hypothesis is supported
6. Replication: other scientists repeat the experiment to check the results
7. Theory/law: if the hypothesis is consistently supported, it becomes a theory or law
Key Features of Science (conventional view):
- Objectivity: the scientist's personal values do not influence the findings
- Empiricism: knowledge comes from observable, measurable evidence
- Falsifiability: theories can be tested and potentially disproven
- Replicability: experiments can be repeated by other scientists
- Causation: science identifies cause-and-effect relationships
- Generalisability: findings can be applied beyond the specific case studied
- Cumulative knowledge: science builds on previous findings, progressively improving understanding
Positivism: Sociology CAN and SHOULD Be a Science
Positivists argue that sociology should adopt the methods and logic of the natural sciences.
Key Arguments
- Durkheim (1895) — The Rules of Sociological Method:
- Social facts (institutions, norms, values, statistics) are external to individuals and exercise constraint over their behaviour — they are as real and objective as the facts studied by natural scientists
- Sociology should study social facts using the methods of natural science: observation, measurement, correlation, and the search for cause-and-effect laws
- His study of Suicide (1897) demonstrated this approach: using official statistics to discover social patterns and causal relationships
- Comte (1842): sociology is the "queen of the sciences" — it applies scientific methods to the most complex subject matter (human society)
- Sociology CAN identify laws of social behaviour: just as physics has laws of motion, sociology can discover laws of social change, stratification, and organisation
Positivist Methods
- Quantitative data: surveys, questionnaires, structured interviews, experiments, official statistics
- Large, representative samples
- Standardised, replicable procedures
- Statistical analysis to identify correlations and causal relationships
Evaluation of the Positivist Position
| Strength | Weakness |
|---|---|
| Has produced valuable research: Durkheim's Suicide, large-scale surveys | Interpretivists: human beings are fundamentally different from natural phenomena — they have consciousness, meaning, and free will; they cannot be studied like chemicals or planets |
| Quantitative data allows identification of patterns and trends | Popper: sociology cannot produce falsifiable predictions — it is too imprecise |
| Replicable methods allow findings to be checked | Social facts are not as objective as claimed — statistics are socially constructed (labelling, decisions about what to count) |
| Policy-relevant: governments use survey data for policy decisions | Reductionist: reduces complex human behaviour to numbers — loses meaning and context |
Interpretivism: Sociology CANNOT Be a Science (and Should Not Try)
Interpretivists argue that the subject matter of sociology (meaningful human action) is fundamentally different from that of natural science, and therefore requires a different approach.
Key Arguments
- Weber: sociology must pursue verstehen (empathetic understanding) — grasping the subjective meanings actors attach to their actions
- Human beings are not objects that respond mechanically to external forces — they interpret their situations and act based on those interpretations
- The same "stimulus" (e.g., unemployment) produces different responses in different people because people attach different meanings to their situation
- Science seeks to explain behaviour through external causes; sociology must understand it through internal meanings
Why Sociology Cannot Be Scientific (Interpretivist View)
| Scientific Feature | Problem for Sociology |
|---|---|
| Controlled experiments | Cannot control all variables in social life; ethical constraints prevent manipulation of human subjects |
| Objectivity | The sociologist is part of the social world they study — complete objectivity is impossible |
| Prediction | Human behaviour is unpredictable because people have free will and can change their minds |
| Replication | Social situations are unique — no two interactions are identical; qualitative research cannot be exactly replicated |
| Laws | Human behaviour is too complex and variable for universal laws |
| Measurement | Meanings, emotions, and motives cannot be quantified without losing their essence |
Evaluation
- Strength: Recognises the unique nature of human beings as active, meaning-creating agents
- Weakness: If sociology cannot produce generalisable knowledge, what use is it? Purely descriptive accounts of individual meanings may not advance sociological understanding
- Weakness: Extreme interpretivism leads to relativism — if all knowledge is subjective, no claim to truth is better than any other
Popper: Falsificationism
Karl Popper (1934, 1959) — The Logic of Scientific Discovery:
- Science does NOT work by proving theories true (verification) — it works by attempting to disprove them (falsification)
- A theory is scientific only if it is falsifiable — it must make predictions that could potentially be shown to be wrong
- Scientists should try their hardest to refute their theories — theories that survive rigorous attempts at falsification are provisionally accepted (but never proven)
- Induction (generalising from observations) is logically flawed — no matter how many white swans you observe, you cannot prove "all swans are white" (a single black swan disproves it)
Popper on Sociology
- Marxism is not scientific because it is not falsifiable:
- When capitalism did not collapse as predicted, Marxists reinterpreted the theory rather than abandoning it
- Any evidence against Marxism is explained away ("false consciousness" explains why workers don't revolt) — this makes it unfalsifiable
- A theory that explains everything explains nothing
- Similarly, Freudian psychoanalysis is unfalsifiable — any behaviour can be "explained" by unconscious motives
- Some sociology can be scientific IF it produces falsifiable hypotheses — e.g., Durkheim's claim that suicide rates are higher among Protestants than Catholics can be tested
Evaluation of Popper
| Strength | Weakness |
|---|---|
| Provides a clear criterion for distinguishing science from non-science | Kuhn: this is not how science actually works in practice — scientists do not constantly try to disprove their theories |
| Rightly criticises unfalsifiable theories (Marxism, psychoanalysis) | Lakatos: scientists should not abandon a theory because of one falsification — they should try to save it by modifying auxiliary hypotheses |
| His criterion of falsifiability is widely accepted in the philosophy of science | Too strict — even natural sciences do not always meet this standard perfectly |
Kuhn (1962): Scientific Revolutions
Thomas Kuhn — The Structure of Scientific Revolutions:
- Science does NOT progress through gradual accumulation of knowledge (the conventional view) — it progresses through revolutions
- Normal science: most of the time, scientists work within an accepted paradigm — a shared framework of concepts, theories, methods, and assumptions (e.g., Newtonian physics)
- During normal science, scientists solve puzzles within the paradigm — they do not question the paradigm itself
- Anomalies (findings that do not fit) are ignored, explained away, or shelved
- Scientific revolution: when anomalies accumulate to a crisis point, the old paradigm is overthrown and replaced by a new one (a paradigm shift)
- Example: the shift from Newtonian physics to Einstein's relativity; from creationism to Darwinian evolution
- The new paradigm is incommensurable with the old — they cannot be directly compared because they use different concepts and assumptions
- Science is therefore a social activity — paradigms are maintained by scientific communities, not by pure logic
Kuhn on Sociology
- Sociology is pre-paradigmatic — it has no single accepted paradigm
- Functionalism, Marxism, feminism, and interactionism compete without any one achieving dominance
- Until sociology achieves a shared paradigm, it cannot be a mature science
- Evaluation: some argue this is a strength — competing perspectives keep sociology open and critical; a single paradigm would close down debate
Evaluation of Kuhn
| Strength | Weakness |
|---|---|
| Shows science is a social activity — scientists are influenced by their communities | Popper: this makes science irrational — paradigm shifts are not based on evidence but on social factors |
| Explains why scientific change is revolutionary rather than gradual | Difficult to apply to social sciences — what would a sociological "paradigm shift" look like? |
| Reveals the conservative nature of normal science — anomalies are ignored | Lakatos: oversimplifies — science has competing research programmes, not single paradigms |
Realism: A Middle Ground
Realism (Bhaskar, 1975; Pawson, 1989; Sayer, 1992) offers a position between positivism and interpretivism.
Key Arguments
- There IS an objective reality (agreeing with positivism) — social structures are real even though they cannot always be directly observed
- But the social world is an open system — unlike the closed systems of laboratory experiments, social phenomena are affected by many interacting variables that cannot be controlled
- Science works by identifying underlying structures and mechanisms that produce observable events — not all of these are directly visible
- Example: you cannot see a magnetic field, but you can observe its effects; similarly, you cannot see "class structure," but you can observe its effects on life chances
- Both quantitative and qualitative methods can be scientific — the key is whether they identify real causal mechanisms
Evaluation
- Strength: provides a sophisticated middle ground — sociology CAN study real structures without claiming to replicate lab conditions
- Strength: avoids the extremes of positivism (everything is measurable) and interpretivism (nothing is objective)
- Weakness: the concept of "unobservable structures" is difficult to test — how do we know they are real if we cannot observe them?
Summary Table
| Position | Can Sociology Be a Science? | Key Thinker | Key Argument |
|---|---|---|---|
| Positivism | Yes — should use natural science methods | Durkheim, Comte | Social facts are objective and can be studied scientifically |
| Interpretivism | No — human meaning requires different methods | Weber | Verstehen, not measurement, is needed |
| Popper | Only if it produces falsifiable hypotheses | Popper | Marxism and Freudianism are not science |
| Kuhn | Not yet — no shared paradigm | Kuhn | Sociology is pre-paradigmatic |
| Realism | Yes — but as an open system | Bhaskar | Underlying structures are real but not always observable |
| Postmodernism | No — and science itself is just one narrative | Lyotard | All knowledge claims are relative |
Exam Tip
This topic is worth 20 marks on Paper 3. Structure your essay as a debate: start with the positivist claim that sociology can be scientific, challenge it with Popper (falsificationism) and Kuhn (paradigms), then introduce interpretivism as an alternative and realism as a middle ground. The strongest answers include postmodernism as a radical challenge to the entire debate — if all knowledge is relative, the question "can sociology be a science?" is itself meaningless. However, evaluate postmodernism by noting its self-contradiction (it claims as an objective truth that there are no objective truths).