Race and the Death Penalty in Britain

Identity in the Archives: problems of identification


Since the project launched in April 2017, we’ve been focused on the first four tasks listed in our project plan: identifying cases, collecting data, first reading of data, literature review. These activities are operating concurrently, to some extent. This blog post focuses on the first activity, describing what we are doing to identify relevant cases, how we are doing it, and some of the problems we have encountered so far. As described in our section ‘about the project‘ which is based on the project proposal, we are seeking to identify all cases of black, Asian and other minority ethnic (BME/BAME) defendants who were sentenced to death in the period 1900-1965 to allow us to try to answer our research questions.

The primary problem is:

How do we know which defendants were BME/BAME?

Identifying ‘race’ from names, descriptions and images are all highly problematic in their own ways. And as historians have generally accepted that ‘race’ is historically (and geographically) contingent, racialised categories might change over the 65 years covered by the project. So in criminal cases where the death sentence was pronounced, how do we know who was considered racially ‘other’ (that is, by the predominantly white police and judiciary), when and why? This is important because we have already identified that black and other minority ethnic defendants were more likely to be capitally punished than white defendants (roughly 5% of all executions were of BAME defendants, whereas they represented less than 1% of the total population), which at least suggests that racism(s) was/were in operation in the criminal justice system in the period. But if that was the case, against whom did discriminatory practices operate? How was racial or ethnic difference identified and understood by contemporaries? For example; were lines drawn because of colour, parentage, place of birth, migration status or a combination of these factors? Were there degrees of difference, or were all racial ‘others’ treated the same, across the period of the project, and at different locations in England and Wales? Could a defendant describe their own identity or was it defined for them? …These are just some of the questions we have, and this blog post won’t be able to resolve them but it will describe some of the problems we’ve had with attempting to answer them so far. In so doing, it also describes the various methods researchers might employ to identify pertinent cases in the archives, applying their own research interests or finding criteria.

Since our project is about ‘Race and the Death Penalty’, we also need to find cases where the defendant was sentenced to death. We have a few sources we can call on, including The National Archives catalogue, lists of people hanged published on the internet, contemporary calendars or similar lists of cases or convicted persons and publications about crime, and contemporary newspaper reports.

The National Archives catalogue contains a wealth of information, despite the majority of entries being only a tiny (and sometimes frustratingly thin) representation of what you might find when you actually order the documents at the archives and look in the boxes. I should stress here that the documents within the twentieth century crime files held at the National Archives have not been digitised, let alone their textual content made searchable. So let me be clear that what we are searching is the file descriptions in the catalogue entries – the metadata recorded by archivists that represents a box or a folder sitting on the shelves, the contents of which can only be accessed by going to the National Archives in person, ordering it and physically opening the box.

The National Archives collections that relate to crime (that’s CRIM, ASSI, MEPO, HO, PCOM, DPP and J) contain nearly a million entries. And some file descriptions in themselves can help us to work out who was sentenced to death, as shown in the example below:

National Archives catalogue entry showing a case where defendant was ‘sentenced to death’.

However, The National Archives catalogue can be difficult to navigate and hard to get the results you want from its interface and search function (currently called ‘Discovery’ though I understand a new version is in development). An alternative way of finding information is to download all of the catalogue metadata from the API (Application Programming Interface) and search through it yourself, designing your own queries. Some computer programming knowledge is required for this, and properly describing this method of exploring and selecting cases requires a whole blog post in itself (which I’m still meaning to get around to writing). But file descriptions in the catalogue are not consistent across the whole of the period, nor are they consistent across different court jurisdictions in England and Wales. So we might be able to work out how many files exist in the archive that represent cases where the death sentence was pronounced, but that might not be an accurate representation of all the people who were sentenced to death in England and Wales in that period.

There are also websites like Capital Punishment UK which list all the people hanged in England and Wales in the 20th century. The first problem with this source is that it only gives those actually capitally punished, not those who successfully appealed, or were reprieved, or died in prison. The image below shows an extract listing one year’s hanged:

Screenshot from Capital Punishment UK showing English and Welsh executions in the year 1900.

The second major problem, which relates to all the methods of identifying cases so far mentioned including the National Archives catalogue descriptions, is that these sources might give us the names of defendants sentenced to death, but names don’t tell us very much, if anything, about ‘race’. They might help us to identify some BME/BAME defendants using a rudimentary understanding of ‘foreign sounding’ names, but this is reductive and problematic to say the least. For example, it completely disregards the fact that England and Wales in the past, as now, was made up of a variety of races and ethnicities and so a foreign or unusual name was not an accurate marker of race.

Contemporary publications about crime go a bit further, including descriptions of the people involved and a brief outline of the case. A good example is A Calendar of Murder by Terence Morris and Louis Blom-Cooper (1964) which Lizzie and I have each used in our previous research to identify potentially useful cases. Each entry names both defendant and victim and describes them, briefly, where the authors felt it was important. Unfortunately, the book only covers the period 1957-1964 and it has no earlier equivalent. An example entry from the book is shown below:

Detail from T Morris and L. Blom-Cooper, A Calendar of Murder: Criminal Homicide in England since 1957 (London: Michael Joseph, 1964), p. 239.

Based on Lizzie’s initial research when she wrote the proposal for the project, we suspected that the most consistent and reliable method for identifying cases might be mining newspaper descriptions. This means using the search term “sentenced to death” limited to the period 1900-65 and literally going through every page of results on The British Newspaper Archive (BNA) to pick out the ones where the person so sentenced was described by their racial ‘otherness’ in the headline or first couple of sentences of the article. BNA then allows us to bookmark or download the page so we can use the name and date to search The National Archives (TNA) for the relevant files on the case.

The problem with this approach is the number of results! So far, we’ve checked about 7,000 headlines, but the total is well into the tens of thousands. That’s not indicative of the number of cases since multiple newspaper titles are represented by the British Newspaper Archive, and cases were often reported on in the same newspaper on multiple occasions as they followed the case from start to finish (although obviously it’s only from sentencing to the conclusion of the case that the phrase ‘sentenced to death’ is likely to have appeared). I can’t even do the Maths on how many hours it would take me to go through all the results.

So at the same time as going through the ‘sentenced to death’ results we’ve been using some of the keywords to check that we have collected all the cases where defendants are referred to with this descriptor. A trigger warning here – some of the words that have come up to describe people are pretty shocking. We’ve been surprised by the sheer number of different ways used to refer to the race of a defendant in the past. They include explicit references to colour (eg. ‘Negro’, ‘Coloured’ and ‘man of colour’), which could apparently signify that the person described was black or of mixed ethnic background.  ‘Chinaman’ or ‘Chinese’ was a common referent but could apparently describe defendants from a large area of South East Asia not limited to China proper, including Hong Kong. ‘Celestial’ was also used to describe people from China, and was one I hadn’t heard before (shown in the example below). Other referents to national belonging that explicitly meant racially other were geographical (eg. ‘Indian’, ‘Burmese’, ‘Jamaican’ and ‘West Indian’) and some were occupational: ‘Lascar’ could apparently mean various racial identities and ethnic origins, but specifically referred to men working on ships in certain roles.


Snapshot from British Newspaper Archive search result showing defendant Lock Ah Tam described as a ‘Celestial’ and ‘Chinaman’ in newspaper The Sunday Post.

It could be argued that where there are photographs of defendants (in newspapers or in case files) we can make assumptions about whether a person would have been considered racially ‘other’ but this has its limitations too. For example, Udham Singh, described as a ‘man of colour’ was born in the British Punjab in 1899, held a British passport, and described himself as Indian. However, in the interwar period he travelled the world as a sailor under the name Frank Brazil, listing himself on his identity documents as Puerto Rican because Indian men were not allowed to work on American ships. (More on Udham Singh in an earlier post on our site, here.) Hierarchies of race and colour existed that complicated the identification of an individual by the shade of their skin. This included class connotations as well as specificities of ‘race’.

And significantly, identities could be contested or negotiated, within certain limits. For example, the first case of an arrest with the use of an ‘identikit’. The image produced was black and white, and not necessarily explicitly raced in itself, although the descriptions it came with were (the composite image described the person it depicted as “Indian”). There was some disagreement in court as to whether the image produced by the identikit really looked like the person police arrested based on the identikit image. Had the police officer who questioned Eddie Bush detained him because he resembled the image he had seen, or had he identified him because he was told to be on the lookout for ‘an Indian youth and his blonde girlfriend’ in the Bloomsbury area? (She was apparently very striking looking.) It was a matter of debate whether Eddie even ‘looked Indian’. Born in London, Eddie’s mother was white and his father was from Pakistan. Eddie himself didn’t identify as Indian/Pakistani or a man of colour, and resisted attempts to characterise him as such. In fact, in one narrative of the murder he said that the victim had called him a ‘n****r’ which had made him so angry he had been provoked to attack her.



Photograph of Eddie Bush, left. Identikit used by police to identify Bush, right.


With all this in mind, the question remains, whose definitions of race are important to our project?

We don’t have the answer yet, but a recent breakthrough has some exciting implications. We noticed that some of the Home Office folders within the files had little notations on the back of them. Examples are shown below:


Three examples of notes on Home Office files reading ‘C.M.III. – Chron. List’ etc.

They all say “C.M.III. – Chron. List” followed by page numbers and descriptions, the most fascinating being “p. 230 – Foreigners” and “p. 231 – coloured men.” We wondered, did this refer to an index being kept of all the cases that fitted these (and other) categories by Home Office definitions? It took a long time to work out what C.M.III might mean but we think we’ve got it. Catalogued among the ‘precedent files’ in the Home Office collections at the National Archives are ‘black books’ previously kept in the Home Office itself. It seems that these were referred to for making reports on cases with common elements, elements from murder weapons used (all cases of shooting listed together, for example) to relationships between the victim and defendant (pages and pages of lists, and subsequent reports, of murder by ‘sweethearts’ for instance) and from defence strategies employed (‘provocation’ for example) to the perceived racial identity of the defendant. Much like the published book A Calendar of Murder described above, the lists in these books give the name of the defendant, the year of the trial, and the outcome of the case. Helpfully, they also provide reference numbers for the relevant Home Office files (struck out when the file was destroyed). Some cases appear in more than one list. Crucially, if the Home Office took the decision to reprieve rather than hang a defendant found guilty of murder, a brief explanation of why they did so is given. In many cases, the defendant’s release date and ultimate fate is given.

‘Black book’ extract showing ‘Murder by Coloured Persons (Cont’d)’

This is really fascinating stuff. Historians and Historical Criminologists who deal with any of the types of crime described by these indices (a list of some of the categories shown below – please feel free to contact us for more details) can’t fail to recognise the utility of such lists. Although these categories are imperfect they are a great starting point for us. There seem to be a few omissions and inconsistencies (defendants we would have expected to be considered ‘foreign’ not appearing on that list for example, or one defendant from China being listed as ‘coloured’ while another from the same country is a ‘foreigner’). And undoubtedly these designations of people’s identities are problematic, even arbitrary. But these pages are a great starting point for our project. With the Home Office having arguably the most influence over whether or not a capitally sentenced defendant was hanged or pardoned (their sentence reduced to live imprisonment), literally whether someone lived or died, they provide a very specific but immeasurably valuable perspective from which to view race and racialisation in relation to the death penalty.

A selection of headings under which capital cases were categorised in Home Office black books.


Watch this space for a list of all cases of black, Asian and minority ethnic people who were sentenced to death in England and Wales in the twentieth century, including an exploration of the Home Office’s classifications.