Geographic coordinates are not exact, but a best-effort interpration of historical data

The Willmott 1799 Census of Carrick-on-Suir: A Structured Interpretation

The Willmott Census of 1799 is one of the most detailed surviving snapshots of everyday life in Carrick-on-Suir at the end of the eighteenth century. Unlike later censuses, it records not only household heads but also children, servants, lodgers, and occupational detail, offering a rare view of the social and economic structure of an Irish provincial town on the eve of the nineteenth century. This project presents a structured, searchable, and spatially explorable interpretation of that material, designed to make the census accessible while remaining faithful to its historical character.

The original census does not survive as a single, unified dataset. Instead, it exists as a collection of interrelated tables extracted from a legacy DataEase application, separating households, people, servants, children, marital records, and household segments. These tables were internally consistent but not relational in the modern database sense. As part of this work, the source files were carefully mapped into a coherent data model, linking people to households, households to places, and places to approximate locations within Carrick-on-Suir, while preserving the provenance of every record.

Several interpretive decisions were required. Households were treated as the primary structural unit, reflecting how the census was originally compiled. Duplicate rows arising from overlapping source modules were deduplicated conservatively, favouring richer descriptions and clearer role assignments while retaining the ability to inspect raw source rows where needed. Locations were resolved manually to street- or area-level centroids rather than precise buildings, acknowledging both the limitations of eighteenth-century addressing and the need to avoid false cartographic precision.

The result is not a modern census, nor a claim of exact address-level accuracy, but a historically grounded exploration tool. It is intended to support research, interpretation, and curiosity—allowing users to move fluidly between people, households, streets, and the wider spatial fabric of the town—while remaining transparent about the compromises and editorial judgements required to bring an eighteenth-century document into a twenty-first-century digital form.

About the views

The Explorer offers two complementary ways of viewing the data. “See as recorded” presents the material in a source-faithful form, where each entry reflects a single row from the original extracted records, preserving spelling, abbreviations, and structure as they appear in the historical source. “Modern view” applies limited, transparent interpretation to improve readability and navigation: clearly identical records may be grouped, household members organised by role, and fuller forms preferred where supported by multiple source rows. No new facts are introduced in this process, and all underlying records remain accessible for inspection. The two views are therefore not different datasets, but different presentations of the same evidence, allowing both close source criticism and practical exploration.

 

Note on occupational spelling and truncation

A small number of occupational terms in the source files exhibit truncation artefacts, most commonly the loss of an initial letter (for example, “CHOOLMASTER” for “SCHOOLMASTER” or “ROCER” for “GROCER”). These artefacts originate in the legacy extraction process and are present in the raw source tables.

In this interpretation, such cases have been corrected at display level only, where the intended meaning is unambiguous, in order to preserve semantic clarity for readers. The original raw values are retained in the underlying data and remain accessible for inspection. This approach reflects an editorial correction rather than a historical alteration, and is intended to improve readability without obscuring provenance.

DataEase ZIP archive contents

The downloadable archive
willmott1799_dataease_extraction_pack_v8_unmarried_transform.zip
contains the following structured material:

Documentation

  • README.md
    Overview of the extraction and transformation process.

  • 00_manifest/files_manifest.csv
    A complete inventory of included files.

  • 00_manifest/application_map.md
    Mapping notes describing how legacy DataEase tables relate to modern concepts.

Extracted relational tables

Located under 02_extracted_tables/, organised by source module:

  • HOUSHAAB

    • households.csv

    • people_in_households.csv

    • segments_by_house_ref.csv
      Core household structure and membership.

  • CHILHAAA

    • children_people.csv

    • segments_by_house_ref.csv

    • CHILHAAA_tokens.csv
      Child-specific census entries and annotations.

  • SERVHAAA

    • servants_people.csv

    • segments_by_house_ref.csv
      Servant records linked to households.

  • MARRHAAB

    • marriage_segments.csv

    • segments_by_house_ref.csv

    • MARRHAAB_tokens.csv
      Marital and relationship-related segments.

  • _ascii_extracts/
    Plain-text dumps of selected legacy tables for reference and verification.

Original legacy application files

Under 01_raw_app/, the archive preserves the original DataEase artefacts (.DBA, .DBM, .DBR, .CAT, .Ixx), including files such as:

  • ADDEHAAA.*

  • CHILHAAA.*

  • TRADHAAA.*

  • UNMAHAAB.DBR

  • CATALOG.CAT

These files are included unchanged, to allow future researchers to revisit the original application-level structure and assumptions.

A pdf guide to the data is available here.