Many British buildings are adorned with plaques, marking the birthplace or residence of a famous person, or the site of a significant event. Details of these plaques are available in an online database, and I thought it would be interesting to see how many of them have a musical connection.

The entire database of British plaques can be downloaded as a JSON or CSV file from openplaques.org.¹ The data includes, among other things, the following details for each plaque…

• a unique id number
• the date it was erected
• its location (latitude and longitude, as well as the street address and geographical region)
• the colour of the plaque
• its subject(s)
• the full inscription
• the organisation responsible for it (local councils, societies, etc)
• links to photographs and to other information on the location or subjects

The database, when I downloaded it, listed over 14,000 plaques across the UK. Although they are often known as “blue plaques”, only 49% of them are listed as being blue: about 10% are black, 7% green, 6% bronze, plus a range of other options including stone (4%), maroon (0.3%) and clear (0.04%).

This map shows the location of all of the plaques (apart from a handful without location data, and some with coordinates in other parts of the world despite being marked as British). As the points are transparent, a darker red indicates overlapping points – i.e. several plaques close together. 22% of plaques are in London, almost ten times the number in Edinburgh (2.3%) which is in second place, just ahead of Birmingham, Liverpool and, surprisingly, Norwich.² There are even a few plaques in the sea – some are on shipwrecks!

Turning to the question of music-related plaques, the problem is how to identify whether a plaque has a musical connection. After quite a lot of trial and error, I identified the music plaques by searching the inscriptions for one or more of the following terms…

"anthem", "_arranger", "_band[slm]*", "blues", "_choir[sm]*", "_chor[ai]", "clarinetist", "_compose", "_concert[os]*", "_conductor", "_gig_", "guitar", "_hymn", "jazz", "_lyric[si]", "_mus_", "music", "operas*_", "orchestra", "_organ_", "_organist_", "_pianist", "recording", "reggae", "singer", "_song", "symphony", "_tenor", "trumpet", "_tune", "_viol[ai]", "_vocal"

An underscore (_) indicates a “word boundary” – i.e. the start or end of a word. Square brackets indicate one of the characters contained within them. An asterisk indicates zero or more of the preceding character. Thus "blues" will be matched by “blues” or “darkbluesky”, whereas "_organ_", with the word boundaries specified, will only be matched by “organ”, and not by “organise”. "_choir[sm]*" will be matched by “choir”, “choirs” or “choirmaster”.³

The process for developing this list was to run the test, examine the output to check for those wrongly classified, and then adjust the definitions and test again. I may have missed a few, and wrongly included others, but there is probably no perfect solution.

This approach forces us to think about what we mean by a music-related plaque. Although most examples are straightforward, here are some more questionable inscriptions that were classified as music-related…

“Here lived Scientist and Musician Sir William Herschel 1738-1822 from where he found the planet Uranus, March 13th 1781 he also discovered Infrared radiation in 1800 ~ and his sister Caroline Herschel early woman scientist 1750-1848 Hunter of comets” (#881: Bath)

“100+ dog years. Music critic, dog poet, photographic model and all round good egg, Barking Lord Scruff of Highgate, lived here, 1985 – Nov ’99. Erected by good friends.” (#12817: London)

“1859-1907 Francis Thompson Poet was born in this house Dec 16 1859. ‘Ever and anon a trumpet sounds / From the bird battlements of Eternity'” (#9201: Preston)

“Theatre Lane. In 1795 the Theatre Royal, seating 600, was created in Marischal Street by Stephen Kemble, brother of the actor John Philip Kemble. Eminent performers included Charles Macready and Charles Keen. The theatre flourished until 1872 when it was replaced by Her Majesty’s Opera House, later the Tivoli, in Guild Street.” (#3548: Aberdeen)

“Jazz Carlin. These steps have been named ‘Carlin Steps’ to celebrate Jazz Carlin’s achievement in winning 2 silver medals for swimming at the Rio Olympics 2016” (#44676: Bradford on Avon)

William Herschel was primarily known as an astronomer, although it seems reasonable to include him for his musical activity as well. The inclusion of Barking Lord Scruff, being a dog, is perhaps more debatable. Francis Thompson’s plaque merely has a musical reference in the quotation used. The fourth example is a building that was replaced by an opera house, so only has an indirect musical connection. Jazz Carlin, apart from her name (actually Jazmin), does not have a musical connection at all.

There are several more examples of these tenuous cases. Short of reviewing them all individually, it is hard to see how an automated process could avoid a few such discrepancies. Overall, 646 plaques were identified as having a musical connection. Whilst this number is small enough to review manually, it would, of course, also be necessary to check the remaining 13,000+ for cases that were wrongly rejected.

The database includes links to further information about the subjects of each plaque, categorised according to type – man, woman, animal, place, group, or thing. Among the 646 music-related plaques there are 936 subjects. Of these, 63% are men, 9% are women, 9% are places, 8% are groups, 1% are things (including ships, events, the “Vox” amplifier, and the “Magical Mystery Tour”), and 0.3% are animals (all dogs), with the remaining 10% being unclassified.⁴

The further information also specifies the subjects’ roles. Among the men, 25% were composers, 13% band members, 9% musicians, 5% singers, and 2% conductors. The women were 20% singers, 10% music hall artistes, 6% band members, and 3% composers. The places were mainly performing spaces, with the most common category (11%) being recording studios. Groups were 84% bands, followed by choirs (4%), clubs (3%), and orchestras (3%).⁵

Several subjects are commemorated by more than one plaque. Here are those with four or more:

• Edward Elgar (15 plaques)
• John Lennon (8)
• Franz Liszt (6)
• Gustav Holst (5)
• Benjamin Britten (5)
• The Who (5)
• Eric Coates (4)
• Jimi Hendrix (4)
• John Ireland (4)
• Marie Lloyd (4)
• Paul McCartney (4)

The following chart shows the distribution of year of birth for the individual men and women commemorated on music-related plaques. There are no women subjects who were born before 1750. The peak period for women is the end of the nineteenth century (singers and music hall stars), whereas the men peak in the second quarter of the twentieth century (jazz and popular music performers).⁶

Interestingly, 62% of music-related plaques are blue, compared to 49% of all plaques. The map of music-related plaques, like the map of all plaques shown above, shows a reasonable spread across the country, although the concentration on London is even higher (30% of music plaques, vs 22% overall). The next highest numbers are in Liverpool (2.3%), Leeds, Newcastle upon Tyne, Gateshead, Manchester, Bristol and Aberdeen (1.4%). Among places with at least 100 plaques, Belfast has the highest proportion of music-related plaques (7.4%), and Oxford the lowest (0.7%).

The following map shows the location of music-related plaques in London. The densest area is around the central theatre district and Soho, but there are other concentrations, for example around Marylebone, St John’s Wood (the location of Abbey Road Studios) and Kensington, that would warrant further investigation.⁷

There is more that can be done with this data – my aim has been to illustrate some of the methodological issues and the sorts of questions that can be addressed. The use of maps is a huge subject in its own right, which I will return to in a future article.

1. CSV (comma separated values) is a text format often used for spreadsheet-like rectangular tables of data. JSON (JavaScript Object Notation) is a more complex format that can cope with messy data, such as nested tables and irregular-length entries, and is probably preferable for this dataset.
2. Norwich seems to have a particularly large number of plaques on its historic buildings, compared to other towns and cities.
3. These strings are examples of REGEX, or “regular expressions” – a standard method of working with text data on computers, although note that the normal REGEX for a word boundary is \b: I have used _ here for clarity.
4. Not all of the 936 subjects had a musical connection – such as Caroline Herschel in the example above. Around 15% of subjects had a role that was not matched by any of the list of musical terms used to check the inscriptions (above), although several of these, on inspection, do have other musical connotations.
5. If more than one role was mentioned, such as “composer and conductor” the first has been used here, on the basis that it is probably the most significant.
6. The chart suggests that women born around the end of the twentieth century may have gained slightly relative to men from the same period. However, the numbers involved are too small for this to be statistically significant.
7. This rather pretty base map is “watercolor” produced by Stamen Design from OpenStreetMap data – see http://maps.stamen.com/watercolor.