Scope and Branches of Anthropology

History, development, aim and scope. Broca founded the Anthropological Society of Paris (1859); the Royal Anthropological Institute dates from 1871. Evolutionists such as Tylor (Primitive Culture, 1871) and Morgan (Ancient Society, 1877) ranked societies into stages. Tylor defined culture as "that complex whole" of knowledge, belief, art, morals, law and custom acquired as a member of society, and became the first Reader in Anthropology at Oxford (1884). The Torres Strait expedition (1898: Haddon, Rivers, Seligman) began field-based work; Boas built American four-field anthropology; Malinowski's Argonauts of the Western Pacific and Radcliffe-Brown's The Andaman Islanders both appeared in 1922. In India, the Asiatic Society dates from 1784 and the Anthropological Survey of India from 1945. Hallmarks: holism, comparison, relativism.

Relationship with other sciences. Anthropology shares human biology with anatomy, genetics and medicine, and society with sociology; Radcliffe-Brown called social anthropology comparative sociology. It shares the past with history and palaeontology, and mind with psychology. Evans-Pritchard (Marett Lecture, 1950) placed social anthropology nearer history and the humanities than the natural sciences.

Branches and their interrelationship. The four fields are biological (physical) anthropology (primatology, palaeoanthropology, genetics, growth, variation), archaeological anthropology, social-cultural anthropology (ethnography describes, ethnology compares) and linguistic anthropology; applied anthropology is often a fifth. Linguistic anthropology spans descriptive linguistics (phoneme: smallest sound unit that changes meaning; morpheme: smallest unit of meaning; syntax), historical linguistics (language families; Swadesh's glottochronology dates divergence), sociolinguistics, and Hymes's ethnography of speaking. The Sapir–Whorf hypothesis says language shapes thought (strong form determinism, weak form relativity). Birdwhistell named kinesics and E.T. Hall proxemics. The fields interlock: language, genes and artefacts together trace migrations.

Epistemology, ontology, theoretical perspectives; research types and hypothesis. Ontology asks what reality is; epistemology asks how we can know it (objectivism, constructionism, subjectivism). In Crotty's scheme (1998), epistemology informs a theoretical perspective (positivism, interpretivism, critical inquiry, feminism), which shapes a methodology (ethnography, grounded theory, survey), which selects methods (interview, observation). Quantitative research measures and tests deductively; qualitative research interprets meaning inductively; mixed methods combine them. Designs are exploratory, descriptive, explanatory or experimental. A hypothesis is a testable statement about variables; rejecting a true null (H0) is a Type I error (α), retaining a false one a Type II error (β).

Fieldwork tradition, ethnography, observation and interview. Frazer (The Golden Bough) typifies the armchair tradition. Malinowski made long participant observation in the local language the standard, to grasp the native's point of view. Ethnography is both process and written product; Geertz took thick description from Gilbert Ryle; Pike coined emic (insider) and etic (observer). Observation is participant, non-participant or quasi-participant; interviews are structured, unstructured or in-depth, often with key informants.

Case study, life history, focus group, genealogy, schedule. A case study probes one unit holistically. A life history shows culture through one biography, as in Oscar Lewis's The Children of Sanchez (1961). A focus group is a moderated discussion in a small homogeneous group, descended from Merton's focused interview. Rivers's genealogical method (1900, from the Torres Strait work, later the Todas) collects pedigrees to recover kin terms, descent, marriage rules, migration and vital statistics: triangle male, circle female, = marriage, vertical line descent. The respondent fills a questionnaire (literacy needed); the investigator fills a schedule face to face.

PRA, RRA and grounded theory. Rapid Rural Appraisal (late 1970s–80s) let outsiders learn fast through semi-structured interviews and observation, and was largely extractive. Participatory Rural Appraisal, associated with Robert Chambers, hands analysis to the community: social and resource maps, transect walks, seasonal calendars, Venn (chapati) diagrams, matrix and wealth ranking. Grounded theory (Glaser and Strauss, 1967) builds theory from data by constant comparison and theoretical sampling until saturation; coding runs open, axial, selective (the last two from Strauss and Corbin).

Exploration, excavation and GIS. Exploration (survey) locates sites by maps, field walking, surface collection and remote sensing. Vertical excavation (trenches) reveals the cultural sequence; horizontal (open-area) excavation exposes one occupation level; Wheeler's grid method keeps baulks that show the section. Stratigraphy rests on superposition (lower is older unless disturbed), and each find's three-dimensional provenience is recorded. A GIS stores georeferenced layers as vector (points, lines, polygons) or raster (grid cells) data for site distribution and predictive modelling.

Variables, sampling, central tendency and dispersion. Variables are independent or dependent, discrete or continuous, on nominal, ordinal, interval (no true zero) or ratio (true zero) scales. Probability sampling: simple random, systematic (every kth unit, k = N/n), stratified, cluster, multistage. Non-probability: purposive, quota, convenience, snowball. Mean suits interval or ratio data, median ordinal, mode nominal; for moderate skew, mode ≈ 3 median − 2 mean. Dispersion: range, quartile deviation (Q3 − Q1)/2, mean deviation, SD and CV = SD/mean × 100.

Parametric and non-parametric tests; regression. Parametric tests assume interval data and normality: t-test (two means), ANOVA's F (three or more), Pearson's r. Non-parametric counterparts: Mann–Whitney U (Wilcoxon signed-rank if paired), Kruskal–Wallis, Spearman's rho; chi-square tests association of nominal variables. These are bivariate; multiple and logistic regression are multivariate. Linear regression fits Y = a + bX by least squares: b = Σ(x − x̄)(y − ȳ)/Σ(x − x̄)², a = ȳ − b x̄; R² is the variance explained. Logistic regression handles a binary outcome, modelling ln[p/(1 − p)] = b0 + b1X by maximum likelihood; e^b is the odds ratio.

Content, discourse and narrative analysis. Content analysis (Berelson) is the objective, systematic, usually quantitative description of manifest content: code units into categories and count. Discourse analysis studies how language in use constructs reality and power (Foucault; critical discourse analysis, Fairclough). Narrative analysis reads sequence, plot and voice; Labov's model runs abstract, orientation, complicating action, evaluation, resolution, coda.