You know the scene.
Perhaps it’s in a 1930s cop movie, or maybe it’s a 1970s Cold War thriller. In either case, there might well be a moment where a bespectacled drone leads our hero into a huge, dark basement. The lights flicker on, illuminating the cobwebs in the corners. Rats scurry across the floor. Ahead of the hero: a vast bank of wooden drawers. His heart sinks, for he knows that somewhere within the interminable contents of those drawers will be the single, minute, piece of evidence which will prove the guilt of the gang boss or the identity of the traitor. Or maybe it’s a 1960s private detective thriller, where our hero arrives to see a blood-spattered body on the floor, surrounded by hundreds of scattered pieces of card, and knows at once that the one bearing the crucial clue has been stolen by the killer.
…and that, dear reader, was how we used to do historical research in the days before databases and Google. Yes, welcome to the world of the card index. The world that was once mine, and in one sense, still is.

It’s difficult now to conceive of just how ubiquitous the card index was. In a nutshell, pretty much everything that would now be stored on a database had to be fitted onto small pieces of blank card and stored in a suitable receptacle. Such an index was only as good as the people who conceived it, the system they devised, and the durability of said receptacle. I once proved the latter in spectacular fashion at the John Rylands University Library, Manchester, where the catalogue was on a card index in very large, sturdy looking wooden drawers. But they were not quite as sturdy as they seemed to be; pulling on one (‘C’, if you must know) with what I thought was only modest force, the whole thing jumped at me like a ravenous lion, with cards scattering to all corners and the drawer itself falling to the floor with a crash that probably did for several of the older and more somnolent readers.
Undeterred by this calamity, I created my own miniature version. When I began my doctoral research on the officers and men of the Restoration Navy in 1982, I realised pretty quickly that I needed a detailed index of all the captains and lieutenants of the period, which as much biographical information as I could muster on them, to enable me to carry out comparisons of, say, social origin and career structure. My starting point was Pepys’ register of sea officers, evidently compiled around the time he left office in 1689 and printed in volume 1 of the Calendar of the Pepysian Manuscripts at Magdalene College, Cambridge. So I produced cards for every officer on the list, about 1,500 men in all: two to a card in the cases of officers with very brief careers, one per card for those with many commissions and/or relatively famous careers. Onto each, I wrote in longhand the details from the Pepys list, usually just the post held (name of ship, lieutenant or captain), the year of each commission, and the name of the person who signed the commission; for the years 1660-73, for example, this was invariably James, Duke of York, the Lord High Admiral. Then, as I went through other sources over the years, I added extra information to the cards.
It quickly became clear that Pepys’ list had many inadequacies. Dates were sometimes simply wrong, or else confused; so, too, was the identification of people, particularly those with similar names. So gradually, my own index began to become much more accurate than any single source on which it was based. A few examples, chosen from many:
- Pepys listed one officer called Peter Belbin, and allocated him lieutenants’ berths on the Rupert in 1672, the Gloucester in 1673, the command of the Sweepstakes in 1673, and then the post of first lieutenant of the Mountague in 1677. But according to ADM10/15 at the National Archives – a very similar source to the Pepys list, but which gives exact dates of service evidently drawn from information in the original ships’ pay books (long since lost), and every single entry and date in which I again added longhand to the card index (!) – there were actually two Peter Belbins, father and son, with the father holding the first three posts (albeit in the order Gloucester first, then Rupert) and the son having the commission on the Mountague. From other sources, I discovered that Peter senior was 63 in 1678, when he was superannuated on the grounds that he was too old to hold further office at sea; a Portsmouth man, he had also been the master of a number of important warships for at least twenty years, including the First Rate St Michael.
- Pepys listed three John Hubbards, two of whom were commanding ships at exactly the same time. He gave ‘John I’ seven commands, ending with the Falcon in 1670, and ‘John II’ eight, ending with the Assistance in 1668, and noting of ‘John II’ that he was ‘slain in fight with some Algier men-of-war in the Streights, 1668’. But it was actually ‘John I’ who was killed in battle, when in command of the Falcon, in November 1669; ‘John II’s command of the Assistance actually began on 1 January 1671, and he died in command of her in the West Indies in July 1671. So ‘John II’ had the longer career, the opposite of what the Pepys list suggests.
- Pepys shows one John Wood, captain of four small ships from 1660 to 1667, second lieutenant of the St Andrew in 1672, captain of the Kent in the same year, then lieutenant of five ships in 1673-4 and of three more in 1676-81. But again, these were two different men: ‘John I’ was dismissed the service after being held responsible for the wrecking of the Kent in October 1672, while ‘John II’s last four commissions were actually as captain, with three of them being large and prestigious frigate commands. So relying on Pepys alone would give a completely inaccurate picture of the careers of these men.
But it wasn’t just a case of sorting out cases of mistaken identity in the Pepys list, or rectifying the significant number of omissions. Often, I was able to add detail that made the men in question real, living people, rather than just names on a page. For example, a quick skim of the Pepys list would suggest that Captain Argenton Alington had a brief and unremarkable career, serving only as lieutenant of the Charles in 1668 and captain of the Guernsey in 1669 before being ‘slain in fight with Algier men-of-war off — in the Streights 166-‘. In fact, Alington was killed on 3 July 1670, and his death was greatly mourned. He was the brother of the third Baron Alington, MP for Cambridge, and it was said of him that he was ‘a gentleman greatly to be lamented, as being a person of exceeding promising hopes’; Lord Alington was immensely proud of his brother’s career and his heroic death, even though he knew that it might well mean the end of his family’s male line and with it, the title. Then there was Thomas Penrose, recorded in the Pepys list with a bare entry showing his command of the Monck during the Second Anglo-Dutch war. But from other evidence, Penrose was clearly a colourful character – a client of the ship’s namesake, General George Monck, Duke of Albemarle, he was a Cornishman who kept his wife aboard his ship during the winter of 1665-6, by which time it was said that he ‘grows debauched’ and was much addicted to drink. (Hmm, now there’s a thought: I think Thomas Penrose really ought to put in an appearance somewhere in the Quinton Journals!)
I might well try and track down some other interesting information from the card index for future posts. In the meantime, though, I really must see about getting the material transferred into a database…after all, the world is full of criminal gangs desperate to get their hands on the exact dates of each commission held by, say, Lieutenant Endimion Drake (no relation – or was he?), or to sort out which Captain John Johnson was actually which. You can’t be too careful, after all.