Family Reconstitution
- Introduced by
- Louis Henry and Michel Fleury; refined by E. A. Wrigley and R. S. Schofield (Cambridge Group)
- Year
- 1956
- Type
- Nominative record-linkage and demographic estimation
- Family
- Nominative historical demography
What does it do?
Family reconstitution is the nominative technique that rebuilds the demographic experience of historical communities by linking individual baptism, marriage and burial entries from parish registers into the reproductive and mortality histories of identifiable families.
Developed by Louis Henry and Michel Fleury in 1950s France and brought to its fullest expression by the Cambridge Group for the History of Population and Social Structure under Wrigley and Schofield, the method transforms an undifferentiated stream of vital events into Family Reconstitution Forms, each documenting a marriage, the births of children, and the deaths of family members. From these forms the analyst derives age-specific fertility, age at marriage, birth intervals, infant and child mortality, and adult survivorship. Because it works at the individual level it can isolate demographic behaviours—such as spacing or stopping—invisible to aggregate counts, making it the gold standard for pre-census population history.
The intuition behind the method
Imagine a parish's registers as a shuffled deck of index cards, each recording one event for one person on one day. Alone, a card tells you little; a single burial says nothing about how long that person lived. Family reconstitution sorts the deck by stitching cards that belong to the same person and the same household back together. A marriage card opens a folder; the baptisms of children, the burials of infants, and eventually the burials of the spouses are slipped into it. Once a couple's folder is complete you know when the wife married, how many years she was exposed to childbearing, when each child arrived, and when the union ended. Multiply this across hundreds of folders and the parish's hidden demographic regime appears: how early people married, how widely births were spaced, how many infants died. The genius is that exposure—time genuinely at risk—can be measured, so rates rest on proper denominators rather than guesswork.
History
The technique was formalised by Louis Henry and Michel Fleury at France's Institut national d'etudes demographiques in 1956, growing out of Henry's interest in natural fertility and his Geneva bourgeoisie studies. It crossed the Channel through the Cambridge Group for the History of Population and Social Structure, founded in 1964, where E. A. Wrigley and R. S. Schofield refined linkage rules and exposure accounting, culminating in their monumental reconstitution of twenty-six English parishes published as English Population History from Family Reconstitution (1997). Continental work by Knodel on Germany and the broader European Fertility Project extended the approach, embedding family reconstitution as the foundational nominative method of modern historical demography before computerised record linkage broadened its reach.
When to use it
Use family reconstitution when continuous, kin-rich parish or church registers survive for a stable community over several generations and you need individual-level demographic measures—age at marriage, marital fertility schedules, birth spacing, infant and child mortality—rather than mere population totals. It is the method of choice for testing hypotheses about natural fertility, deliberate family limitation, or the demographic response to crises in pre-statistical populations. It is appropriate where registration is reliable and migration is modest enough that exposure can be tracked. Avoid it where records are fragmentary, naming is too homogeneous to permit confident linkage, or the population is so mobile that most families slip out of observation before their histories complete.
Limitations and considerations
- Extremely labour-intensive, restricting coverage to a handful of well-documented parishes
- Migration removes families from observation, truncating and potentially biasing histories
- Results from atypical record-keeping parishes may not represent the wider population
- Linkage is unreliable where surnames are homogeneous or registers have gaps
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Yields true age-specific rates from measured person-years exposure, not guessed denominators
- Reveals individual-level behaviour such as birth spacing and stopping invisible to aggregates
- Allows internally consistent comparison of fertility and mortality across parishes and eras
- Provides the demographic micro-foundations for testing natural-fertility and family-limitation theories
Applications
- Measuring the rise and decline of natural fertility in pre-industrial English parishes
- Detecting deliberate family limitation through birth-interval and stopping behaviour
- Estimating infant and child mortality before the registration era
- Studying age at first marriage and the European Marriage Pattern
- Reconstructing demographic responses to subsistence crises and epidemics
How it works
How it works
Every entry in the baptism, marriage and burial registers is transcribed into a standardised record capturing names, dates, event type and any stated kin relationships such as father, mother or spouse. Spelling variants, Latin forms and abbreviations are normalised. This corpus of atomised events is the raw material; its completeness and the richness of recorded kin information set an upper bound on what reconstitution can later recover from the community.
Formula
E = \{e_i = (\text{name}, \text{date}, \text{type}, \text{kin})\}Events are matched into chains referring to the same individuals using names, dates and kinship constraints. A baptism naming given parents is attached to that couple's marriage; a burial is tied to the person whose vital span it plausibly closes. Henry's rules demand consistency: ages, birth order and intervals must cohere. Ambiguous links—common surnames, remarriages—are flagged or excluded rather than forced, preserving the reliability of the surviving linkages.
Formula
L(e_i, e_j) = 1 \iff \text{name}_i \approx \text{name}_j \wedge \text{kin consistent}Successful links are assembled onto a Family Reconstitution Form, one per marriage, recording the wedding date, each child's baptism and any burial, and the deaths of the spouses. The form is the analytic unit: it makes a couple's complete reproductive and survival history legible. Forms are graded by completeness, since fertility analysis requires knowing when observation of the wife began and ended.
Formula
\text{FRF} = (m_x, b_{x_1}, \dots, d_x) \text{ per couple}For each single year of age the method sums the time couples and individuals were genuinely under observation in the parish—present, not yet dead, not yet emigrated. This person-years denominator is the heart of the technique's rigour: rather than dividing events by a guessed population, it divides by measured exposure, so that periods of in- and out-migration are handled by truncating observation rather than biasing the rate.
Formula
E_x = \sum_c \int_{x}^{x+1} \mathbb{1}[\text{couple } c \text{ in observation}]\,dtAge-specific marital fertility is the ratio of births to wives' person-years at each age; mortality quotients divide deaths by exposure in the relevant age band. Because numerator and denominator are drawn from the same linked population, the resulting schedules of fertility and mortality are internally consistent and directly comparable across parishes and periods, yielding the natural-fertility curves and mortality levels that define historical demography.
Formula
f_x = \frac{B_x}{E_x}, \quad q_x = \frac{D_x}{E_x}Raw rates are adjusted for known biases: under-registration of infant deaths and baptisms, the loss of families to migration, and the truncation of incomplete forms. Wrigley and Schofield devised weighting and selection procedures—such as restricting fertility analysis to families observed for their whole reproductive span—to keep estimates unbiased, and stress sensitivity testing so that conclusions do not hinge on any single correction assumption.
Formula
\hat{f}_x = f_x \cdot w_{\text{migration}} \cdot w_{\text{underreg}}Common pitfalls
- Treating incomplete reconstitution forms as if they captured full reproductive spans
- Ignoring infant-death under-registration, which inflates apparent early-life survival
- Forcing ambiguous links instead of excluding them, contaminating the linked population
- Failing to distinguish out-migration from death when an individual disappears from records
Frequently asked questions
How does it differ from aggregative analysis?
Aggregative analysis counts undifferentiated baptisms, marriages and burials month by month to track trends, whereas family reconstitution links events to specific individuals and households. Only reconstitution can compute age-specific rates and reveal behaviours like birth spacing, because it knows who was at risk and for how long; aggregation never identifies individuals or measures exposure.
Why is migration such a problem?
The method assumes it can observe families across their reproductive lives. When a couple emigrates, their later births and deaths occur elsewhere and vanish from the registers, truncating the history. Unless out-migration is distinguished from death, exposure is mismeasured and rates are biased. Restricting analysis to families demonstrably present throughout the relevant span is the standard, if costly, remedy.
Can the results represent a whole country?
Cautiously. Reconstituted parishes are selected for exceptional record quality and stability, so they may differ from the mobile or poorly documented majority. Wrigley and Schofield combined many parishes and cross-checked against aggregative and back-projection estimates to argue for national representativeness, but extrapolation from a small set of communities always demands sensitivity analysis and triangulation.
Related methods
Sources
Wrigley, E. A., Davies, R. S., Oeppen, J. E., & Schofield, R. S. (1997). English Population History from Family Reconstitution 1580-1837. Cambridge University Press.
ISBN: 9780521590150Copied.Wrigley, E. A. (1966). Family Limitation in Pre-Industrial England. The Economic History Review, 19(1), 82-109.
Open DOI referenceCopied.Wrigley, E. A., & Schofield, R. S. (1981). The Population History of England 1541-1871: A Reconstruction. Edward Arnold / Harvard University Press.
ISBN: 9780674690073Copied.