Skip to content
Back to search results

Serial History

Introduced by
Francois Furet; Pierre Chaunu; Ernest Labrousse
Year
1971
Type
quantitative descriptive method
Family
serial analysis
Font size

What does it do?

Serial history, histoire serielle, is the historical method that takes as its object not the unique event but the series: a long, homogeneous sequence of comparable facts, such as grain prices, baptisms, burials, marriages, notarial acts, or wages, recorded at regular intervals.

Theorized above all by Francois Furet and practiced by Pierre Chaunu, Ernest Labrousse, and others in the orbit of the Annales school, it grew from Fernand Braudel's call to attend to the slow rhythms of history and from the conviction that the proper data of history are repeated, measurable facts rather than singular happenings. By constructing such series and analyzing their movements, trends, cycles, and fluctuations, the serial historian reconstructs the economic and demographic structures and conjunctures within which events occur. The decisive methodological requirement is homogeneity: the units must be defined and measured consistently across the whole span, so that change in the numbers reflects change in reality rather than in the recording.

The intuition behind the method

A single harvest failure is an event; the movement of grain prices across two centuries is a structure. Serial history is the decision to study the second rather than the first. The insight is that many historical sources were generated repeatedly and uniformly by institutions, parishes recording every baptism, markets posting prices, notaries registering each contract, so that, gathered up, they form long sequences that can be counted and charted. Once you have such a series, history looks different. You can see the slow secular trend, the recurring cycle, the seasonal rhythm, and the sharp crisis, layered on top of one another, much as Braudel distinguished the fast time of events, the medium time of conjunctures, and the almost motionless longue duree. The price of this power is a strict requirement: every entry in the series must mean the same thing. If a parish changed how it recorded burials, or a currency was debased, an apparent trend may be an artifact of the records. Homogenizing the data, ensuring like is compared with like across centuries, is the serial historian's central craft.

History

Serial history grew out of the Annales tradition's turn from events to structures. Fernand Braudel's 1958 essay on the longue duree and his Mediterranean distinguished the layered temporalities, fast events, medium conjunctures, slow structures, that serial analysis would track. Ernest Labrousse's studies of prices and the economy under the Old Regime, and Pierre and Huguette Chaunu's vast Seville et l'Atlantique, exemplified the construction of massive series. Francois Furet's programmatic essays, including Quantitative History (1971), gave the method its name and rationale, distinguishing serial from merely quantitative history and insisting on the homogeneous series as the unit of analysis. Historical demography, advanced by Louis Henry's family reconstitution, was a major application. From the 1980s the cultural turn and microhistory reacted against serial history's impersonality, though its techniques persist in economic and demographic history.

When to use it

Use serial history when long, continuous runs of homogeneous records survive and the research question concerns structures, trends, and rhythms rather than unique events: the movement of prices and wages, demographic regimes of birth, marriage, and death, the volume of trade, or the frequency of certain acts over time. It is the method of choice for economic and demographic history and for any inquiry that seeks the slow conjunctures and longue-duree structures behind the surface of events. It pairs naturally with archival content analysis, which can generate codable series, and with structural analysis of the long term. It is unsuitable where records are short, broken, or heterogeneous, or where the object of study is precisely the singular, the qualitative, or the experiential.

Limitations and considerations

  • Requires long, continuous, homogeneous runs of records that exist only for some places and facts
  • Homogenizing changing records is laborious and error-prone, and flawed homogenization corrupts results
  • By privileging the repeatable and measurable, it neglects the unique, qualitative, and experiential
  • Series reflect what institutions recorded, embedding the biases and gaps of the registering bodies

Source conditions do not replace independent scientific review or validation by an analysis engine.

Strengths

  • Reveals long-run trends, cycles, and structures invisible in event-centered narrative
  • Treats routinely generated records as systematic data, exploiting their volume and regularity
  • Distinguishes layers of historical time, separating slow structures from conjunctures and shocks
  • Provides a quantitative, comparable basis for economic and demographic history across long spans

Applications

  • Reconstructing long-run movements of grain and commodity prices and real wages
  • Charting demographic regimes through series of baptisms, marriages, and burials
  • Measuring the volume and rhythm of trade from port and customs records
  • Tracking the frequency of notarial acts to study credit, property, and family strategies
  • Establishing economic conjunctures and crises behind the events of political history

How it works

How it works

1

The serial historian begins by identifying and constructing a series: a long run of a repeated, comparable fact drawn from continuous institutional records. Candidate facts are those that were registered routinely and uniformly, such as baptisms, marriages, burials, grain and commodity prices, rents, wages, or notarial acts. The historian gathers these from the surviving registers and assembles them into a chronological sequence. The value of the enterprise depends on the length and continuity of the run, since structures and slow movements only become visible over long spans.

Formula
\{ y_t \}_{t=1}^{T}, \quad y_t \ \text{homogeneous, comparable across } t
2

Because real records change over time, the historian must homogenize the series so that each entry is genuinely comparable. This means holding definitions, units, and coverage constant: converting currencies and measures to a common standard, adjusting for changes in administrative boundaries or recording practice, and correcting for gaps and under-registration. Homogenization is the most demanding and consequential step, for without it apparent movements may merely reflect shifts in how the data were produced. The historian documents every adjustment so the constructed series remains transparent and criticizable.

Formula
\tilde{y}_t = \mathrm{normalize}(y_t), \quad \text{constant definition, units}
3

The historian then analyzes the movement of the series, decomposing it into components: a long-run trend, medium-term cycles or conjunctures, seasonal rhythms, and short-term fluctuations or shocks. Techniques range from simple moving averages and index numbers to graphical and statistical decomposition. This separation distinguishes the slow structural drift from recurrent cyclical movement and from the sharp deviations marking crises, allowing each layer of historical time to be examined on its own and related to the others.

Formula
y_t = \mathrm{trend}_t + \mathrm{cycle}_t + \mathrm{season}_t + \varepsilon_t
4

Interpreting the decomposed series, the historian reads its movements at Braudel's distinct temporal levels: the rapid time of events, the medium time of the conjuncture, and the slow time of the longue duree. Turning points, accelerations, and crises in the series are correlated with one another and with known historical developments, asking how demographic and economic rhythms framed the events traditional narrative foregrounds. The aim is to grasp the structures and conjunctures that conditioned human action rather than to chronicle the singular.

Formula
\Delta y_t, \ \text{movements at } \{ \text{event}, \text{conjuncture}, \text{longue duree} \}
5

Finally the historian interprets the patterns substantively, always in light of how the underlying archive was produced. Movements in the series are read as evidence about past economic, demographic, or social reality, but the interpretation remains alert to the homogeneity assumption: a trend is only as trustworthy as the constancy of its recording. The serial historian thus integrates statistical pattern with source criticism, treating the numbers as traces of both the past and the institutions that registered it, and drawing structural conclusions accordingly.

Formula
\mathrm{structure} = g(\{ y_t \}) \mid \text{archival homogeneity}

Common pitfalls

  • Mistaking a change in recording practice for a real change in the phenomenon measured
  • Comparing units that are not truly homogeneous, such as unconverted currencies or shifting categories
  • Ignoring under-registration and survival bias that distort the apparent level and trend
  • Reifying statistical artifacts into historical structures without corroborating evidence

Frequently asked questions

What distinguishes serial history from quantitative history generally?

Furet drew the line at the series. Quantitative history applies numbers to the past in many ways; serial history specifically constructs long, homogeneous runs of a repeated, comparable fact and studies the movement of the series itself as the primary object. The unit of analysis is the sequence over time, not the individual event or a one-off statistical snapshot. Homogeneity across the whole span is the defining methodological requirement.

Why is homogeneity of the series so important?

Because the method infers real historical change from movement in the numbers, every entry must mean the same thing across the entire span. If recording practices, currencies, measures, or administrative boundaries shifted, an apparent trend may be an artifact of the records rather than the past. The serial historian must therefore convert units to a common standard and adjust for changes in coverage, documenting each correction so that movements reflect reality, not the recording apparatus.

How does serial history relate to Braudel's longue duree?

Closely. Braudel distinguished three layers of historical time: the rapid time of events, the medium time of conjunctures, and the slow, almost motionless time of structures, the longue duree. Serial history provides the technique for observing these layers, since decomposing a long series into trend, cycle, and short-term fluctuation makes the different temporalities measurable. Serial analysis is thus the empirical instrument for the structural history Braudel called for.

Related methods

Sources

  1. Furet, F. (1971). Le quantitatif en histoire. In J. Le Goff & P. Nora (Eds.), Faire de l'histoire (Vol. 1, pp. 42-61). Gallimard.

    ISBN: 9782070287666
    Copied.
  2. Braudel, F. (1958). Histoire et sciences sociales: La longue duree. Annales. Economies, Societes, Civilisations, 13(4), 725-753.

    Open DOI reference
    Copied.