¹ This project has been financially supported by the Faculty of Arts at the University of Helsinki and National Library of Finland and Digitalia in Mikkeli. Leo Lahti has been partially supported by Academy of Finland (grant 256950). We would like to express our gratitude also to Kaius Sinnemäki, Hege Roivainen, Maija Paavolainen and Krister Lindén for their help in various different ways. An earlier version of this paper was presented at the Liber 2015 conference in London. We would like to thank our audience, and especially Marian Lefferts of CERL.

² For more about the ESTC (and the historical development of the catalogue and the difference between the English and the Eighteenth-Century Short Title Catalogues), see http://estc.ucr.edu/estcdean.html including the bibliography noted there.

³ This article is planned as the first in a series of publications that will extend to the analysis of European knowledge production in general based on the ecosystem emanating from our work on different library catalogues.

⁴ This initiative for this paper came from the agenda outlined by Michael Suarez (2003–2004), and especially what he states about the use of numbers and tools such as the ESTC (including the limitations) on p. 166.

⁵ See, e.g., ‘Finding the Text: Enumerative and Systematic Bibliography’, in Greetham (1992), pp. 13–46 and ‘Finding Materials’ and ‘Libraries’, in Altick and Fenstermaker (1993), pp. 155–204.

⁶ https://github.com/rOpenGov/estc. We would like to thank the British Library for providing us with the data used in this article.

⁷ We are simultaneously working on many different library catalogues in an open data project. For more information on the research data and the latest advances, see: https://github.com/rOpenGov (estc, fennica, kungliga).

⁸ For a good analysis and example of how to begin to use the ESTC catalogue in the manner that we develop here, see also http://douglasduhaime.com/blog/mapping-the-early-english-book-trade.

⁹ On the development of the ESTC, see Snyder and Smith (2003) and Alston (2004).

¹⁰ For a text-mining project on David Hume’s History of England, see http://www.crassh.cam.ac.uk/events/25988.

¹¹ Veylit (1994) is perhaps the most comprehensive undertaking thus far. At the same time, it is quite telling that although there is a chapter on the digital book in The Cambridge Companion to the History of the Book, first published in 2015, there is no reference to the statistical (or quantitative) approach in the index.

¹² See especially Eliot (2002). For Eliot’s own work on quantitative analysis, see Eliot (1994, 1997, 1998). For more optimism within the community of people taking a quantitative approach to book history, see e.g., Bell (2001).

¹³ There are plans to update the ESTC in the form of ESTC21, with improved linked data features and richer data; see https://estc21.wordpress.com/collecting-data/.

¹⁴ On the relevance of descriptive bibliographies, see Suarez (2015).

¹⁵https://github.com/rOpenGov/estc..

¹⁶ On authorship in general, see Belanger (1982), Myers and Harris (1983) and Lindenbaum (1995).

¹⁷ Of ancient authors, Flavius Josephus (37 – c. 100) and Caesar, Julius (100 BC – 44 BC) would have made the list of most published titles in ESTC, but they have been left out of this study for the sake of focusing on early modern authors.

¹⁸ On the essential and growing amount of scholarship on female authors, see e.g., Summit (2000), Crawford (1984), Ferguson (1996), Lewalski (1993), Mann (1999) and Stanton (1988).

¹⁹ Darby Tracy is a pseudonym, we have left her in the figure to highlight that there are many similar cases in the data.

²⁰ The role of the Stationers’ Company is obviously of the highest relevance in any analysis of London publishing: see Myers and Harris (1997) and Greg (1956, 1966). On provincial book production, see especially Hinks (2012). It is also noticeable that, other than the Latin trade, the number of British books printed on the continent was very high throughout the early-modern era. On the relationship between British books and the continent, see also Hoftijzer (2002).

²¹ On early modern US printing, see Martin (2007) and Amory and Hall (2000). On Dublin, see Pollard (1987) and Phillips (1998). On Irish provincial printing, see Wheeler (1978), and Gillespie and Hadfield (2006).

²² You can also download the full video: https://github.com/rOpenGov/estc/blob/master/inst/examples/liber.mp4.

²³ See Adams (1980) for a detailed study of British pamphlets and late-eighteenthcentury American disputes, and Giles (2001) for more on the eighteenth-century development of American literature).

²⁴ Some of the data for calculating the paper consumption in ESTC catalogue is missing so that this has an impact on this particular figure. We have estimated that approximately 24% of the information for US paper consumption is missing from this graph. The effect of this is such that the US bar might be up to 32% higher (the missing information does not effect Ireland or Scotland to any significant extent). Nevertheless, the US paper consumption is still considerably lower, even if all the data was available. The updated figures can be accessed on github when the data becomes available: https://github.com/rOpenGov/estc/blob/master/inst/examples/20151023-LIBER.md.

²⁵ In this sense, we are still battling with the same questions as posed in Pocock (1957).

²⁶ For an overview of bibliography, see Greg (1913), Tanselle (1974, 1976–1977, 1981, 1988) and Gaskell (1972).

²⁷ A problem pointed out by Maureen Bell in an unpublished paper delivered at the British Library in 2006.

²⁸ On the use of the London average print run, cf. Raymond (2003) p. 90 and the works cited there.

²⁹ One of the major shortcomings of the ESTC is the lack of data on eighteenthcentury newspapers, which is why we have excluded most of the later newspapers from this study. On the statistical approach to eighteenth-century newspapers, see Aspinall (1948).

³⁰ For general studies on the development of the British book trade in the eighteenth century, see Myers and Harris (1981, 1982), Alston (1984) and Feather (1986).

³¹ Unfortunately we cannot provide a paper-consumption comparison for all the documents in the ESTC because we have not yet obtained access to this data from the British Library.

³² We have been waiting for some time to see if it is possible to access the full ESTC data from the BL to complete this analysis.

³³ On the British paper industry, see Coleman (1958), Hills (1988) and Thomson (1974).

³⁴ On paper and its uses in Britain, see Pollard (1941–1942) and Heawood (1929, 1930a,b, 1931, 1947).

³⁵ This graph clearly also reflects the economic aspects of printing: see e.g., McKenzie (1992).

³⁶ On printing in general during the civil-war era, see Lambert (1981, 1984, 1987) and Frank (1961). One aspect for further study, apart from the effect of duplicates caused by the presence of the Thomason Tracts, concerns the Greek and Latin works imported from the European continent. It is obvious that not all of the ‘Latin trade’ is included in catalogues such as the ESTC. This is not really a problem in our study given that our aim is to focus on book publishing through the ESTC catalogue rather than the absolute objective reality, and thereby to present a reliable view of general trends and patterns. On the ‘Latin trade’, see Roberts (2002).