Since uploading old data and notes to Github has gone well and I’ve got some spare time, I’ve added some notes about sources for tracing the careers of parliamentarian military units and their personnel. This is very rough and nowhere near complete because it’s left over from abandoned projects and things that I noted while looking for something else.

The final Github upload for now is some notes and references about the establisment of the New Model Army’s artillery train. The army was mainly planned by the Army Committee, but the Accounts Committee objected to the establishment for the artillery and produced its own smaller establishment. These lists and the debate between the committees give some useful insights into the intended (which is not necessarily the same as the actual) size and composition of the artillery train, and into how armies were expected to operate. For example:

• Medium field guns (sakers) were not expected to be concentrated in artillery batteries, but to be split into pairs attached to foot regiments or in ‘places of advantage’.
• The role of pioneers included cutting gaps in hedges so that infantry and cavalry could march off the road, freeing up road space for artillery and transport.
• More evidence to support Stephen Bull’s argument that artillery was already using canvas cartridges in the 1640s.

During the British Civil Wars, Parliament appointed committees to collect accounts from each parish of losses and contributions to the parliamentarian war effort, including taxes, sequestration, voluntary contributions, and anything that soldiers had taken without paying. This post is about parish loss accounts from Buckinghamshire, but the accounts from Warwickshire are better known. Although I looked at many of the original Warwickshire accounts for my PhD research in the 1990s, I haven’t had anything to do with them since. Transcripts of most known Warwickshire accounts are now free to view at the Heritage and Culture Warwickshire website (although they’re missing TNA, SP 28/38/3, f. 202 which covers Ipsley). The transcripts are in PDF format, which makes it easy to publish and archive the text, but difficult to extract structured data for systematic analysis. These accounts were transcribed by a team of volunteers, who gained new skills and experience, and engaged more deeply with historical sources. This is good public engagement, which is the proper purpose of crowdsourcing, but crowdsourcing is not necessarily a quick or cheap way to get work done. According to the introduction, it took 26 volunteers 2 years to transcribe around 200 accounts (which I think added up to about 1 million words but I can’t find a reference for that now). Although the volunteers weren’t paid for their labour, the project still needed a grant of £13,800 from the Heritage Lottery Fund. As a professional transcriber, I can transcribe a million words on my own in a year or less if someone pays me to do it. At the time when the Warwickshire project started, I would have accepted £13,800 to be guaranteed work for a year, although I would charge much more now.

I haven’t transcribed the Buckinghamshire accounts, but I have compiled some catalogue data and uploaded it to Github. The files consist of:

• a catalogue of around 100 accounts that survive.
• a list of all known parishes and sub-units in Buckinghamshire with references to surviving accounts, and also to evidence that an account once existed but has been lost. Also a few incidental references to other documents covering the same place, but these are not complete.
• a link table to link IDs in the above two datasets.

There may well be more Buckinghamshire accounts that I haven’t found. The Warwickshire project found some accounts in unexpected places. Based on what I’ve found, 39% of units in Buckinghamshire have surviving accounts, 17% had accounts that have since been lost (although some might turn up somewhere), and 44% have no trace of any accounts. There is a very noticeable gap in Newport hundred, where no accounts are known to survive, and the only trace of a lost account is an individual account for one person. This doesn’t look random, but there’s no definite explanation for it.

The latest upload to my Github is a fairly large dataset about compounding cases. These relate to people whose property was confiscated by Parliament during the British Civil Wars. The system started with the Sequestration Committee, which had the power to confiscate the estates of anyone it classed as a ‘delinquent’. It’s important to note that the official definition of ‘delinquent’ changed more than once, and that it never meant exactly the same thing as the modern word ‘royalist’. For example, someone’s estate could be sequestered for not paying taxes, even if they never did anything for the King. Later, Parliament set up the Committee for Compounding, which had the power to return sequestered estates if the owner paid a fine equivalent to a fraction of the full value of the estate and swore an oath that they wouldn’t help the King in future. Later still, the rules were changed so that delinquents could compound on their own discovery. This meant that anyone worried that they might be classed as a delinquent could confess and pay the fine in advance to avoid possible sequestration in future.

The records of the Committee for Compounding are held by The National Archives of the UK in series SP 23. In the 19th century, Mary Green compiled Calendar of the Committee for Compounding, a published summary of each case with references to original documents. As the calendar has been out of copyright for a long time, I’ve compiled spreadsheets containing structured data about compounding cases, and people who were main subjects of cases. This amounts to over 6,000 cases and over 8,000 people. This was for a project of my own that has now been abandoned, so I might as well share what I’ve got. There was going to be a second pass where I recorded details of military service and which articles of surrender people claimed the benefit of, but that won’t happen now. I hope this data will be useful to someone.

This is about some of the data I uploaded to Github last week, but it deserves a short post of its own. In the 17th century, many business premises were identified by sign boards with a picture painted on them, and were known in speech and writing by a phrase that described the picture, such as ‘The White Bear’. It’s important to remember that these signs could identify almost any kind of premises. It was only later, after other businesses stopped using them, that they came to be particularly associated with pubs. This dataset is a list of 249 of these building signs from addresses in and near London, taken from the accounts of horses contributed on the Propositions. This is a small dataset compared with some of the other data I’ve uploaded, but it’s a fairly large sample of signs compared to the few that can be extracted from other sources I know of, and I think it should be safe enough to treat it as if it’s a random sample. There’s plenty of scope to analyse the iconography of these signs either quantitatively or individually. What were the most popular colours and symbols? What meanings did they have besides identifying the people and businesses that occupied a building? What’s the story behind ‘The Dog’s Head in Pot’, which was the name of not one but two buildings in London?

This week I’ve uploaded to Github a fairly large amount of data and documentation about the Propositions. This was a system of voluntary contributions to the parliamentarian war effort in the early part of the First English Civil War. It was very successful at first but didn’t last long and was replaced by compulsory taxes. The readme file in the main directory that I’ve linked to gives a slightly more detailed overview of the system, with links to sources. There’s also a list of manuscript records that I haven’t transcribed but that could be useful if anyone else wants to research the Propositions in the future.

The most important dataset that I’ve uploaded is a transcript of three account books in TNA, SP 28/131/3-5 listing horses, arms, and riders contributed for the parliamentarian army in 1642 and 1643. These books name thousands of horse owners, often with their address and status. They also give descriptions of the horses, often including details of colours and markings. The transcript is marked up with XML. This data needs more systematic analysis, as most historians have only cherry-picked examples from it. There is also lots of scope for more detailed research on individual people and places. The record linkage that I’ve done is only very rough and could be improved on.

There were probably far more contributions of money and plate than of horses and arms, but there isn’t such a comprehensive record of money and plate. One source that I’ve transcribed and uploaded is an account covering Propositions money and assessments on non-contributors in part of the county of Essex. This can only be a tiny fraction of total contributions, but it’s unique as far as I know in allowing a direct comparison of contributors and non-contributors. Some accounts of money collected in other places are listed in the readme file but not transcribed. Even these are likely to be only fragments of the total.

This week I’ve uploaded to Github transcripts of accounts of money paid out by two of the treasurers of the Earl of Essex’s army in the First Civil War (there was a third treasurer but I haven’t found any accounts of his payments). Together these include over 4,000 rows of data giving details of payments to army officers, providing useful evidence for the careers of individual units and members of their personnel. There may be a few entries for civilian contractors for purchase or hire of goods or services. Like the New Model Army procurement records, these could be useful for business history.

These accounts are strong evidence for some important facts that may not be well-known, and where better known sources are either misleading or are correct but are weaker evidence than contemporary financial records (see readme file for more details and references):

• the parliamentarian forces at the siege of Portsmouth were commanded by Sir John Meldrum, not Sir William Waller.
• Philip Skippon’s foot regiment first appears in the surviving accounts for Essex’s army in November 1642. Skippon was first paid as Major-General of Essex’s army in January 1643. In both cases, there may have been earlier payments from the other treasurer whose accounts have not been found.
• Henry Billingsley was moved sideways from Lieutenant-Colonel of Denzil Holles’s foot regiment to be assistant to the Major-General in late August 1642.
• Nathaniel, John, and Francis Fiennes were all paid only as captains of horse in 1642. Nathaniel was first paid as Colonel in March 1643.

Now that the wiki is out of the way, I’m uploading some separate datasets to Github in case they’re of any use for anyone else’s research. To start with, I’ve added a catalogue of manuscript volumes created by the Ordnance Office in London recording contracts, deliveries, and issues of equipment for the New Model Army from 1645 onwards. This is obviously relevant to military history, and it is potentially very valuable for business history. The names of suppliers and the descriptions of equipment they supplied show what someone was actually trading in. This can be obscure in 17th-century London because women were usually described by their marital status, and men’s occupational descriptors could refer to their actual trade or livery company membership, which were often not the same thing. Matching contracts, deliveries, and warrants and receipts for payment can sometimes reconstruct a whole transaction, showing not only who supplied what, but the time from contract to delivery to payment. Gerald Mungeam published the largest collection of contracts from the London Museum (and scans of the original manuscript are now free to view at their website), but the rest of these records are less well known and have not been used to their full potential for business history.

The wiki has now been deleted. All content is archived and freely reusable at Github. The domain name will redirect to Github for some time. From now on, I will be working on sharing more datasets about the British Civil Wars that were never imported to the wiki. I will post about them here as they appear on Github.

As part of the process of decommissioning By The Sword Linked, I’ve shared at Github the python scripts that I used to generate batches of pages to import to the wiki, along with documentation and some example data. The process involves creating structured data as delimited text files, which you can do with a spreadsheet, Open Refine, a custom script of your own, or any other method. The python scripts provided can turn these files into wikitext XML, which can be imported into MediaWiki. This is a more efficient way of creating large numbers of wiki pages than doing it manually or using a bot. The scripts were designed for By The Sword Linked but could be used for other wikis with few or no changes.