1: Why only British & Northern Irish films?
2: Why setting and not location?
3: Surely not all films represent places to the same degree?
4: How are you working out where films are set?
5: Why film and not TV?
6: Why fiction and not documentary?
7: Why is x film not on the map?
8: Can you add my film?
9: But didn’t you expect London to be over-represented?
Q: Why only British & Northern Irish films?
A: I’m certainly open to expanding the coverage of this map! It grew out of a PhD project to examine the cinematic potential of overlooked parts of England. In order to work out where had been overlooked I needed to plot settings of films on a map. The data I got from the BFI covered films from the whole of the UK, which was a broader coverage than I needed, but which was still just about manageable for a solo researcher.
Any efforts we make to expand the map further will be dependent on other countries’ national cinema boards having similarly comprehensive data holdings to that of the British Film Institute.
Q: Why setting and not location?
A: I’m interested in representation. Consider The Personal History of David Copperfield (2019, Armando Iannucci). A good portion of the story is set in Victorian London. Only small pockets of London look as they would have in Victorian times, and they are expensive and impractical to shut down and control for a film shoot. The production negotiated with the city of Hull to shut down some of its intact historic streets instead. But those streets are playing London; Hull does not play itself. The only viewers of David Copperfield who would feel that Hull is represented by the film are those who are local to the area. To most people, the film would represent London. Similarly, the town of Great Yarmouth is played by King’s Lynn.
This does not just happen in period dramas where the original setting no longer fits the period of the story. Fast & Furious 6 (2013, Justin Lin) could not afford to shut down the streets of London for its car chases. Instead, the film cuts from wide establishing shots of landmarks like the London Eye to stunts filmed on the streets of Liverpool. The illusion is helped by setting the action at night, which allows inconvenient detail to be hidden in darkness, and with judicious use of props, such as replica Corporation of London street furniture and a red double-decker bus (running the number 4 route) which crops up in the background at various points in the film. There are also some non-action shots cut into the sequence which do appear to be shot in the actual City of London.
Q: Surely not all films represent places to the same degree?
A: Absolutely. A film that sells 20 tickets doesn’t represent a setting as well as one that sells several million. Successful films put places on the map. We are going to introduce further layers of visualisation to the map in order to take into account a film’s ‘reach’ or the breadth of representation it confers on a place.
However, it is worth bearing in mind the concept of depth of representation. Films that do bigger business may engage with the places they represent more shallowly, and in a less considered way than smaller films. For example, James Bond films feature exotic locations, but these representations of places play on archetypal (some might say stereotypical) images of the places. The depiction of place has to be digested by the audience at the same time as watching spectacular stunts and as a result Bond films say little about places that the average viewer does not already know. Smaller films that rely less on spectacle and more on character can engage more deeply with place. Red Road (2007, Andrea Arnold) is a psychological thriller with social realist elements, and is named after the Glasgow housing estate (since demolished) in which the drama was set. Measuring ‘authenticity’ of representation is beyond the scope of statistics and maps, and can only be done through more traditional humanities approaches, such as close reading and critical analysis.
Q: How are you working out where films are set?
A: I am cross-referencing various databases. The primary one is a list of films provided by the British Film Institute, which includes fields such as title, date of release, name of director and short plot synopses. This data is a definitive list of British films which received a cinematic release but does not include settings or locations. I am compiling my own list of locations, drawing on various sources. The primary source is the archives of Sight and Sound magazine, which includes concise plot synopses alongside the reviews. Other sources include the trailers of the films and interviews with the films’ directors in local newspapers and industry magazines, which sometimes shed light on the filmmakers’ use of location and setting. In certain situations, often based on synopsis and reading of genre, it was possible to state with a degree of confidence that location and setting in specific films were one and the same, and therefore lists of filming locations on sites like imdb.com could be used to infer settings.
Although in theory I could list every setting in a film, I typically list up to five locations per film. So far in the research I have encountered few films that feature more than five locations for any considerable duration. Duration is important: just as the longer a film stays with a character the more we know about them, the longer a film stays in one location the more rounded the depiction of that location.
Once locations are inferred and added to the database, I cross-reference with a geographical database that assigns a unique code to each settlement and landmark in the UK, and a specific latitude and longitude. This allows me to distinguish between towns which share names, such as the various Newcastles. If you zoom in you will find the ‘pin’ for a city is located in the geographical centre of that city rather than in the area that most residents would consider to be the town centre.
It is worth noting that the map of settings does not accurately depict the number of films made in an area. For example, Northern Soul (2014, Elaine Constantine) is set in a fictional town that is an amalgam of several places in the area of Greater Manchester, and so five dots appear on the map when looking at this setting view. We are developing a different visualisation that more accurately maps how many films have been set in each local authority of the UK.
Q: Why film and not TV?
A: Simply, scope. The feature film is a tidy and defined product and therefore is easier to systematise than a more amorphous form like the TV programme. If I attempt to widen the scope of this map to include TV programmes, there are many categories to consider. Do I include TV movies? Mini-series? Long-running serial drama? Soap operas? Semi-scripted reality shows like The Only Way is Essex? Besides the plethora of content, there is also the issue that documentation of these productions is not kept in one central place the way information about feature films is collated.
Part of this disparity in data collection is historical. Government policies conceive of feature films as high-value commodities for export. The cultural cachet of films (although not regarded as highly in the UK as other forms such as theatre) has traditionally been greater than that of small screen productions. Both of these factors contributed to the establishment of a body to coordinate, document and archive the activities of the film production sector.
Q: Why fiction and not documentary?
A: Again, scope. The PhD research that this map is related to is concerned with the cinematic potential of overlooked places in the UK for fiction film production. In some respects, documentary films may be easier and more straightforward to map. Places in documentaries almost always play themselves. But as a solo researcher, I needed manageable objectives. I would not rule out including documentary films in future versions of the map.
Q: Why is x film not on the map?
A: Firstly, it may not be on the map yet. The BFI data I am working from lists films by their date of release in the UK, not the date of their world premiere. So, to look at a recent example, The Personal History of David Copperfield had its world premiere in at the Toronto Film Festival on the 5th September 2019, but it was released in UK cinemas on the 24th January 2020, and that is the release date in my database. At the time of writing, I have processed films with UK release dates between 1st January 2010 and 31st December 2019.
Secondly, I have excluded films from the map that did not appear to have any scenes set in the UK. It is impossible for me to watch all the films to check, and so some may have been wrongly excluded. A list of films that are not on the map (with reasons for their exclusion) is here. If the film you’re concerned about is on the list, please get in touch to let me know what UK settings feature in the film, with a brief description if possible, and I’ll amend the data.
Thirdly, it may not be in my data. The BFI has records of all the films it knows about. Most producers will register their film with the BFI because doing so is part of qualifying for production tax credit, which essentially allows them to recoup 20% of their UK production spend. But some films that feature the UK may never register with the BFI. For example, a recent trend in Bollywood features is to include a sequence set in an aspirational foreign location, such as the Swiss Alps. A number of Bollywood films have shot in London, and some of these have set up as co-productions or produced the film as a British-registered company. These films, which often include British-Indian characters, and which may engage the services of UK production and post-production facilities count as British. But through the industry grapevine I’ve heard of other Bollywood productions that simply travel to the UK for two weeks, shoot their scenes, and return to Mumbai without ever registering their production with the BFI. In which case, no record would exist.
Microbudget feature film makers in the UK (often first-time filmmakers) may not be aware of the institutional and professional frameworks that exist, and also may not register their films. Likewise, some microbudget features may have played at festivals but never attracted a distribution deal. To the best of my knowledge I include only films that have had a cinematic release.
If you think I have missed a film, please get in touch with the film’s title, the name of its director and the date of its UK cinematic release and I’ll investigate.
Q: Can you add my film?
A: In principle I’d love to, as there are many excellent films that are made completely independently and which deserve a wider audience. However, practically this study limits itself to films that have been registered with the BFI (and if/when it expands its scope, the relevant national cinema institute). As far as I am aware films on the BFI’s list have received a cinematic release (beyond festival screenings and special events) and/or a distribution deal. As a side note, I advise all independent filmmakers making features in the UK to register their productions with the BFI. Doing so helps you qualify for tax credits that make production and post-production cheaper.
Q: But didn’t you expect London to be over-represented?
A: Of course, because it’s the largest city in the country, but what I’m interested in is how over-represented it is. We are working on an overlay that will show films per capita by area. But in the meantime, the table below will give some sense of how great the disparity of representation is between cities/areas. For instance, there are 20 films set in Glasgow, which means that by total number of films it is the second-best represented city in the UK. But if we look at averages, we see that if there were an artificially equal distribution of filmic production based solely on population, Glasgow could reasonably expect to have featured in 8 more films in the 2010-2019 period.
Using this comparison to average films per capita, there will be very few places better-represented than London, but there is at least one. Oxford is 3% better represented than the capital given the number of films set there relative to its size.
|Area||Population||Films||People per Film||Ratio|
|London||8,500,000||307||26,813||0.64 (or roughly one third better than average)|
|Glasgow||1,200,000||20||60,000||1.43 (or roughly 40% worse than average)|
|Oxford||154,000||6||25,666||0.61 (or roughly 40% better than average)|
|West Midlands||5,900,000||13||453,846||10.84 (or roughly 11 times worse than average)|
|Herefordshire||191,000||2||95,500||2.28 (or just over two times worse than average)|
The implementation of the overlay is likely to be more sophisticated than this rough table, and will calculate averages based on administrative areas as well as by towns and cities. One film set in a Herefordshire village of 2,000 people would make that place statistically better represented than London, which contradicts all other evidence. A better approach in this case would be to measure based on the population of the entire county of Herefordshire.
Q: I have another question that you haven’t answered here.
A: Great! Get in touch and ask me and I’ll update this page.