Books and Borrowing 1750-1830


Where can I find details of the original sources and where to find them?

See our About the Database page

I want to cite material from this resource in an article. Can you tell me how to do this?

Each page has a ‘Cite this Page’ link that you can click.

Is recording practice different in different libraries?

Yes. The original borrowers’ registers record information in a variety of different formats and orders. We have standardised this information to make it searchable, but you can see digitised versions of all the original registers using the Image and Text or Image only functions.

Where can I find library catalogues for the collections described?

If a contemporary or modern catalogue exists, you will find details on the Introduction page for each library. The easiest way to navigate there is via the Explore Libraries function.

How can I download data from the site?

Each page has a ‘Download Data (CSV)’ link that you can use to download data from the site. You can also access the project’s API (Application Programming Interface) directly at The API provides all of the pathways to the project’s data that are used by the front-end and allows both computers and humans to query the data and access results in either JSON or CSV formats.

Which library catalogues have you used to derive information?

We have used the extant library catalogue closest in chronological time to the borrowing from the collection that we are recording.

How do you define library types?

In our system, we categorise libraries as Subscription, Institutional or Other. Our ‘Other’ category includes free public lending libraries (Innerpeffray and Haddington Libraries), private libraries (Craigston Castle) and circulating libraries (Chambers Circulating Library). More information is available here: Types of Libraries

What is the difference between a Work, an Edition, a Holding and a Part?

Across the whole of the Books and Borrowing site, we use the following data structure and terminology. All titles in our dataset are categorised on four levels: Works, Editions, Holdings and Volumes. A work is a text considered as a conceptual object or whole (such as Adam Smith’s Inquiry into the Nature and Causes of the Wealth of Nations). An edition is a particular physical printing of a work (such as the 1776 quarto edition of Adam Smith’s An Inquiry into the Nature and Causes of the Wealth of Nations printed in London for William Strahan, Thomas Cadell and William Creech). A holding is a copy of an edition (or in some cases multiple editions, as when pamphlets are bound together) held at one of our libraries (such the copy of the 1776 quarto edition of Adam Smith’s Wealth of Nations held by the Advocates Library). A part is a part of a holding (such as volume 2 of the Advocates Library’s copy of the 1776 Wealth of Nations).

How do you know which edition of a work was actually being borrowed?

In the majority of cases, we have been able to derive this from information in contemporary catalogues or other sources such as accessions lists. Where we have no further information, we have used our knowledge of an individual library’s usual practices, and where this has been impossible, we have assigned the first edition of the work. In all cases, we have given our attribution a confidence level: from certain to speculative. This can be viewed at the top right- hand side of the Edition information.

Can I use this site to find out how many books from an individual library’s collection were actually borrowed?

No. Our statistics cover books borrowed from the library, and we do not have statistics available for what percentage of the full collection this represents. However, if you are interested in finding out about books that were never borrowed, we have provided information (and where possible, links) to contemporary catalogues for each library, where these exist. See the Information tab for each library for more information about catalogues and other contemporary sources.

Why haven’t you transcribed every register for every library?

Constraints of time and funding have meant that we have needed to prioritize transcribing certain registers over others. Our selection principles are outlined here:

Where can I find further information about what you have and have not included in the statistics and visualisations?

You will find this on the Information tab relating to each Library, and in the Facts and Figures pages for each library and for the complete site.

Why do ‘Church of Scotland’, ‘Presbyterian Seceder’, ‘Episcopalian’, Catholic and ‘Non-Presbyterian Dissenter’ appear in the ‘Occupations’ categories? Surely these are religious affiliations, not occupations.

This is correct. These are indeed religious affiliations, not occupations, but in our Occupations categorisation scheme we wanted to differentiate between Ministers/Priests of these different religious affiliations, and our data hierarchy made it impossible to do this without representing these as separate occupations. However, wherever one of these religious affiliations appears, you can assume that the person in question is in fact a Minister/Priest. We have retained the separation out of religious affiliations despite this slight inconvenience, because we think it important for our understanding of the religious and political context of Scotland in the period to keep these differentiations. More information about how we categorised religious occupations can be seen here: Religious Occupations

Why do some borrowers have more than one occupation assigned to them?

This is because they have changed occupation over the course of their presence in the borrowers’ registers. For example, a student who graduates and becomes a librarian will appear both as “Education>Student” AND “Arts and Letters>Librarian”.

What were your principles of categorisation for genre?

Our full categorisation scheme can be viewed here: Documents (‘Genre Classification’). More information about our thinking can be found here:

 What were your principles of categorisation for occupation?

Our full categorisation scheme can be viewed here: Documents (‘Occupations Classification’). We have identified an occupational area (e.g. Law; Agriculture; Religion etc) and then subdivided that into different standardised occupations which reflect both modern terminology and period-specific terms (e.g. Advocate, Writer to the Signet, Clerk of Session; Farmer, Shepherd, Ploughman; Minister, Lay Preacher, Bishop etc).

What were your principles of transcription?

We have used the principles of diplomatic transcription for titles of book holdings and borrowers’ names. We have used standardised forms for all other information.

How do you account for borrowings of multiple volumes of the same work in the statistics?

Our statistics are based on borrowing acts. So if a borrower takes out 3 volumes of a holding at the same time, that would count as one borrowing of the holding. But if the same borrower were to take out each volume separately, at different times, that would count as 3 borrowings of that work. We record both – the number of borrowing records a book holding is associated with (so if multiple book items are taken out at once this is still classed as one borrowing record) and the total number of times each individual book item for a borrowing record was borrowed. This is why on the library’s ‘books’ page we have text such as ‘ Volumes of this book were borrowed 11 times in 7 borrowing records’. For the visualisations, other statistics and the search results we look at the total borrowing records not the total number of book items borrowed, so the lower number.

The Year of Publication search seems to return results that are not published in the year I specified.

Yes, that is correct. This search returns results within date ranges that include your year of publication. For example, a search for “1811” will return a variety of answers, including works published in 1811, but also where 1811 falls within the date range of publication. For example, it will return results for all volumes of the Annual Register because its publication date range is 1759-1837. You can use the bar chart decade and date filters within your search results to narrow further to eliminate these “date range” results.

The Title search seems to return results for other works than the title I searched for.

Yes, that is correct. This search returns results for all edition titles where the word(s) you searched for are present. For example, a search for Pamela will retrieve results for an edition of Sir Charles Grandison “by the author of Pamela and Clarissa”, because those words appear in the edition title. Searching by Book Work in the Complete Search instead will retrieve results for Pamela only.

I’ve found an error in the data. What should I do?

Please contact the project’s PI, Professor Katie Halsey, at, with details of both the error and your correction, and she will ensure the error is corrected.