The project to do work with all the data in the Catalogue of Political and Personal Satires Preserved in the Department of Prints and Drawings in the British Museum has now moved from data wrangling (see part one, part two, and part three – note the community contributions in the comments) to research using that data.

As a footnote to the data wrangling, in the end I used this query…

PREFIX crm: <http://erlangen-crm.org/current/>
PREFIX skos: <http://www.w3.org/2004/02/skos/core#>
PREFIX thes: <http://collection.britishmuseum.org/id/thesauri/>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX bmo: <http://collection.britishmuseum.org/id/ontology/>
PREFIX thesIdentifier: <http://collection.britishmuseum.org/id/>
SELECT DISTINCT ?id (GROUP_CONCAT(?title; SEPARATOR = "|") as ?titles) (GROUP_CONCAT(?name; SEPARATOR = "|") as ?names) (GROUP_CONCAT(?desc; SEPARATOR = "|") as ?descs) (GROUP_CONCAT(?date; SEPARATOR = "|") as ?dates)
    {
        ?object crm:P70i_is_documented_in <http://collection.britishmuseum.org/id/bibliography/294> .
         OPTIONAL {
            ?object crm:P48_has_preferred_identifier ?id }.
           OPTIONAL {
               ?object bmo:PX_physical_description ?desc } .
           OPTIONAL {         
            ?object crm:P108i_was_produced_by ?prodevent.
            ?prodevent crm:P9_consists_of ?prodpart .
            ?prodpart crm:P4_has_time-span ?timespan.
            ?timespan rdfs:label ?date} .
           OPTIONAL {
            ?object crm:P108i_was_produced_by ?prodevent2.
            ?prodevent2 crm:P9_consists_of ?prodpart2 .
            ?prodpart2 crm:P14_carried_out_by ?creator.
            ?creator skos:prefLabel ?name .
            ?creator skos:inScheme thesIdentifier:person-institution}
         OPTIONAL {
            ?object rdfs:label ?title }
    }
GROUP BY ?id

…to extract the data I needed from the British Museum SPARQL Interface. The beauty of this, as Owen Stephens explains, is that not only does it return (see result here) only the values I need – titles, associated names, descriptions, dates – but it concatenates duplicate values into a single string, each separated by the pipe `|` character.

The data still needed a little tidying before I could get going. I pushed the json output into Open Refine then set about regularising the date column. To do this I split the date columns in four using the pipe character as a delimiter (though some entries contain 20+ date, four seemed reasonable having sampled the data). Then taking the first three columns as my date data, I split those again by `-` as many contained values such as `1740-1750`, converted all dates ending in `s` – such as `1740s` – to the middle date (so `1740s` became `1745`, `1750s` became `1755`, and so on) and then took an average of the values that remained (rounding to the nearest integer) to ascertain a smoothed value. There will be some errors caused by this approach, which will come out as I work through the data, but with 23365 lines of data this seemed a reasonable compromise to get at the big picture.

Histograms of the data reveal what we might expect – a gradual increase in the volume of prints in the collection, with 300-400 per year throughout the ‘Golden Age’ of graphic satire (circa 1780-1820). Artist-engravers such as James Gillray and Isaac Cruikshank are prominent within in the collection, especially for prints published in the late-1790s and – in Isaac Cruikshanks case – 1809; a year were he emerged from a period of relative inactivity, boosted no doubt by his then seventeen year old son George Cruikshank, to report in volume on both the spectacular Covent Garden Old Price riots (on which I have written in Nineteenth-Century Theatre and Film and European Comic Art) and the infamous Mary Anne Clarke scandal (which I am currently writing about…). Publishers such as Hannah Humphrey and Samuel Fores are also prominent in the collection, with their decline in single-sheet satirical output as their businesses and the trade diversified from the 1810s visible from the data. And these data tell some useful stories: that the great collaboration of the age, that of Gillray and Humphrey, was indeed close knit, their output one and the same; but also that this collaboration was (probably) an exception to the rule. Fores, with whom Cruikshank published a majority of prints, could go his own way. So too could Isaac (as he did circa 1795-6) and after his death in 1811 the Cruikshank family – now lead by George and his elder brother Isaac Robert, both satiric engravers – forged fresh partnerships, with men such as William Hone, men whose multi-modal publications that contained image and text in greater harmony than hitherto common would come to define the next era of British graphic satire.

In the next post I’ll describe some of the work I’ve been doing analysing the language used to describe the prints in the Catalogue of Political and Personal Satires Preserved in the Department of Prints and Drawings in the British Museum and preliminary investigations into whether the descriptions describe change in the prints or the changing contexts within which those descriptions were made.