‘Texting Scarlatti: Composition, Reception, Performance’ is a two-year research project funded by the Leverhulme Trust (reference number RPG-2022-338), with additional support from the Research England Participatory Research Fund. Although the project is based at the Guildhall School in London, the project team is multi-disciplinary and international: Luisa Morales (Festival Internacional de Música de Tecla Española, Spain), Marco Moiraghi (Italy), Jérémie Lumbroso (USA), Jasper van der Klis (Netherlands) and Barry Ife (UK). A further substantial group of up to fifty volunteer ‘citizen scientists’ is assisting our work.

The project is undertaking the first comprehensive and systematic study of all surviving eighteenth-century manuscript and printed copies of Scarlatti's keyboard sonatas. The words ‘comprehensive’ and ‘systematic’ are crucial to the project's ambitions in terms of scope and methodology:

• • scope, in that the numbers are very large: 555 sonatas in Kirkpatrick's catalogue, plus several dozen others that have been or could be attributed to Scarlatti, preserved in some 3,200 copies, in more than 100 volumes, in over 35 libraries and archives around the world; and

• • methodology, in that the primary objective of the project is to understand how these sources are related to each other, using a combination of philological and phylogenetic techniques. This involves carrying out a note-by-note, bar-by-bar comparison (collation) of each of the copies (‘witnesses’, in the language of textual scholarship) and recording all variants accurately and consistently in both human- and machine-readable form.

Why are we doing this and why has it not been done before? Anyone interested in Scarlatti knows that there is a paradox at the heart of his achievement. Although born and raised in Naples, Scarlatti spent nearly the last forty years of his life in Portugal and Spain teaching one royal pupil in particular: the Infanta, later Queen, Maria Bárbara of Braganza. Yet there is very little documentary evidence of his life and work either inside or outside the court. Only the inventory of household goods drawn up after his death, expertly analysed by Luisa Morales (‘Introducción’ to Domenico Scarlatti en España, ed. Morales (Almería: Asociación Cultural Leal, 2009), 1–11), sheds any light on how he lived and worked. But, somehow, his music went viral across Spain and the rest of Europe. How did this very ‘local’ composer become such a leading figure in eighteenth-century music?

We know something of how his popularity spread in England and France, following the publication of the Essercizi per gravicembalo in London in 1738–1739. And we know more than we used to about how his music circulated in Austria and Germany between the editions of Carl Czerny (1839) and Alessandro Longo (1906–1908). But we know very little about how Scarlatti's music reached beyond the royal music rooms of Portugal and Spain themselves. Who played the sonatas, where, when and how? Who authorized the copying and distribution of scores and how were they put into circulation? These and many other questions have remained largely unanswered for decades. We need to make better use of the huge evidence base we do have – the sonatas themselves – and figure out what they can tell us about musical life in eighteenth-century Iberia and Scarlatti's role in it.

There are several reasons why this work has not previously been tackled at this scale. Some are practical: the sheer weight of evidence has defeated lone scholars in an analogue world, and several of the most interesting sources have come to light since pioneers like Ralph Kirkpatrick and Joel Sheveloff did their research. But the main reason is a major shift of methodological emphasis in textual studies. Whereas classical philology aimed at reconstructing an author's hypothetical urtext, modern philology is more pluralistic, allowing for multiple versions and compositional second thoughts. Shakespeare scholarship has led the way in this regard: the First Folio of 1623 gives us ‘cured and perfect’ texts of thirty-six plays, but it is the individual quarto playbooks that provide the rich depth of performance history. The Arden edition prints three different versions of Hamlet and two of King Lear: different texts, different performances, different plays.

Scarlatti, too, has his ‘first folio’: the Essercizi (k1–30, E) and the two sets of manuscripts now in Venice and Parma (V and P). These are authoritative sources with excellent provenance: Scarlatti clearly had a hand in the former, and V and P once belonged to Maria Bárbara herself. But they are fair copies, and their professional finish can be misleading. Marco Moiraghi's ground-breaking work on the Essercizi (Sonate per clavicembalo, volume 10 (Milan: Ricordi, 2021), reviewed elsewhere in this issue) shows how a source that looks uniform on the surface can be broken into its constituent parts using rigorous philological techniques. And as a result, he, too, prints nine of the thirty sonatas in multiple versions.

‘Texting Scarlatti’, however, is not aiming to produce a new edition of Scarlatti's sonatas in the conventional sense. Kirkpatrick, Kenneth Gilbert and Emilia Fadini (and Eiji Hashimoto – often overlooked) all published good, reliable performing editions based on the most authoritative texts. But a performing edition inevitably offers a single (usually neat and tidy) view of a sonata, whereas the totality of the (usually much messier) evidence provides multiple views. E, V and P account for less than a third of the extant eighteenth-century witnesses, and the question is, how do the more than two thousand others fit in? Our objective is to bring the readings from those other sources more visibly into play so that performers and scholars can evaluate the options open to them. Our work will undoubtedly shed light on a wide range of issues concerning composition and reception, but one thing is certain: every surviving witness records how someone, somewhere in the eighteenth century may have played that piece.

Our comprehensive approach to both the textual and contextual evidence will therefore align Scarlatti more closely with the ‘performative turn’ in twenty-first-century recital practice. By accessing the full range of interpretative options open to them, performers will have greater agency in deciding which pieces to play, how to play them, in which combinations, on which instruments and on the basis of which texts. We aim to open up new ways of programming Scarlatti by focusing not just on the sonatas as objects in themselves, but equally on the collections and their constituent parts that record musical taste within their various locations and performing contexts.

With comprehensiveness comes the need to be systematic. Collation on this scale requires careful attention to detail and skilled judgment. Every marking on a page of music encodes many different kinds of information: pitch, duration, accidentals, ornaments, ties, slurs, stems and beams. A full collation of over 3,200 witnesses will generate about twenty million data points, and each collation must be done at least twice to ensure maximum accuracy. Good-quality data management is essential, and we have therefore developed our own suite of tools, including a highly structured coding system for recording variant readings, and a tailor-made collation platform built by one of our volunteer citizen scientists, Joseph Lou.

This platform, named Roseingrave after Scarlatti's greatest admirer, Thomas Roseingrave (1690/1691–1766), enables all witnesses of each sonata to be collated against a single control text; and because all comparisons are made against the same baseline, we can use the variants they have in common as evidence of kinship. Traditional phylogenetic analysis may no longer be used to establish authorial archetypes, but it remains essential to the task of sorting and organizing the huge number of witnesses for Scarlatti and mapping their interrelationships and circulation.

This work will make an important contribution to the emerging discipline of computational musicology. In due course, we will make our tools and data sets freely available online for other performers and researchers to use. We also intend to deliver a number of specific outcomes from the project:

• • a census of all known eighteenth-century witnesses and their locations in a format that can readily be updated as new sources become available;

• • an index that maps sonatas to witnesses and vice versa, in a format that can be easily accessed and manipulated;

• • a suite of variorum commentaries on each of the sonatas – a full-scale digital appendix to Gilbert's edition (Paris: Heugel, 1971–1984), providing the critical apparatus that has been removed from the online version on the IMSLP website, and serving as a valuable adjunct to other standard editions; and

• • a comprehensive map of the interrelationships between all of the sources with at least some preliminary answers to the questions raised earlier about the circulation and reception of Scarlatti's sonatas across Spain and Europe.

There are several factors that make us confident that we can achieve our objectives. The first is teamwork. The project team contains a range of skills and a variety of backgrounds: a cultural historian and textual scholar with extensive project management experience, three scholar/practitioners with codicological, philological and organological expertise, and a computer scientist. We are particularly pleased to have Jérémie Lumbroso from Princeton University on the team. Lumbroso has built a large repository of digital images of the principal witnesses and has made this resource available to the project, as well as taking the lead on designing and implementing the data architecture.

The project team is also supported by a larger group of volunteers, led by Jasper van der Klis, who are helping us by testing our protocols and procedures, carrying out at least one collation of every witness and giving us a second pair of eyes on everything we do. Their work is supported by the Participatory Research Fund of Research England, which is designed to encourage public engagement and disseminate information about the importance of public investment in research.

As Joel Sheveloff lamented in 1985, compared with other composers of his stature, Scarlatti research is still in its infancy: ‘The popularity of [Scarlatti's] music brought serious study to it earlier than to much other music of the period; we embarked quite a while ago and yet we still cannot see the opposite shore. So many tasks remain that the state of our art would fail aesthetic inspection or information audit’ (Sheveloff, ‘Domenico Scarlatti: Tercentenary Frustrations (Part II)’, The Musical Quarterly 72/1 (1986), 117). We need above all to tackle the challenges of the text before the equally important questions about composition, musical language, performance, organology, dissemination, reception and performance history can be reliably taken further. As W. Dean Sutcliffe puts it in what is the most important study of Scarlatti published so far this century, ‘postmodern musicology can afford to disdain the methods of positivism when so much of the “dirty work” has already been done’ (The Keyboard Sonatas of Domenico Scarlatti and Eighteenth-Century Musical Style (Cambridge: Cambridge University Press, 2003), 7). In Scarlatti's case, much of that unfashionable ‘dirty work’ has still to be undertaken. Our project aims to fill that gap.