Early Music

Editorial

Bringing 'Musicque into the tableture: machine-learning models for polyphonic transcription of 16th-century lute tablature A large corpus of music written in lute tablature, spanning some three-and-a-half centuries, has survived. This music has so far escaped systematic musicological research because of its notational format. Being a practical instruction for the player, tablature reveals very little of the polyphonic structure of the music it encodes—and is therefore relatively inaccessible to non-specialists. Automatic polyphonic transcription into modern music notation can help unlock the corpus to a larger audience, and thus facilitate musicological research. In this study we present four variants of a machine-learning model for voice separation and duration reconstruction in 16th-century lute tablature. These models are intended to form the heart of an interactive system for automatic polyphonic transcription that can assist users in making editions tailored to their own preferences. Additionally, such models can provide new methods for analysing different aspects of polyphonic structure. We have experimented with modelling only voice and modelling voice and duration simultaneously, applying each in a forward- and in a backward-processing approach. The models are evaluated on a dataset containing 15 three- and four-voice intabulations. Each processing approach has its advantages, and the results vary between the models. With accuracy rates between approximately 80 and 90 per cent, both for voice prediction and for duration prediction, the best models’ performance is promising. Even in this early stage of the research, such models yield a useful initial transcription system.

Another lesson from Lassus: using computers to analyse counterpoint The authors report on experiments they have run using the computer to search a small corpus of Renaissance pieces (the famous Lassus duos of 1577) for recurring contrapuntal combinations. They liken these combinations (or ‘modules’ as Jessie Ann Owens has called them) to words in a text, and the process of finding them, to work done by linguists such as John Sinclair on large corpora of text. The program used was devised by a team at McGill University as part of the ELVIS (‘Electronic Locator of Vertical Interval Successions’) project. The interval successions are identified by the vertical intervals and the melodic motions that connect them, in the manner of Tinctoris’s counterpoint treatise (1477), which illustrates most of the possible ways two vertical intervals can be connected. The authors find that some short interval successions appear, as we would expect, in repetitions of thematic material (i.e. as parts of soggetti associated with specific text phrases). Others, however, occur in apparently run-of-the-mill counterpoint: in the middle of words, in the middle of melismas, across phrase boundaries and embellished in a variety of ways. These often exhibit surprising consistency as to semitone position and possible modal associations.

Musical texts and information retrieval: the case of the early modern battaglia Digital technologies can help scholars to navigate the vast quantities of musical data and source materials now available to them, but an imaginative leap is needed in order to conceptualize the kinds of musicological research questions we might ask of electronic corpora. In particular, our data-rich digital world offers enormous potential for the exploration of musical transmission and relatedness. In this article, we explore the 16th- and 17th-century instrumental battaglia (battle piece), a genre with a very distinctive collective identity arising from the use of numerous shared ingredients (including melodic, motivic, textural, harmonic and rhythmic features). However, a battaglia is not defined by the presence of a core set of essential features, and exact concordance between these pieces is often remarkably low. This kind of musical ‘family resemblance’ (formulated after Wittgenstein) poses a serious challenge to both traditional musicological apparatus (for example, finding aids such as thematic catalogues) and Music Information Retrieval (which has often privileged melodic similarity at the expense of other kinds of musical relatedness). This case study provides a stimulus for rethinking the complex nature of musical similarity. In doing this, we outline a set of requirements for digital tools that could support the discovery, exploration and representation of these kinds of relationships.

Partial signatures as transmitted in the Buxheim Organ Book and its concordances In her 1960 dissertation on the Buxheim Organ Book, Eileen Southern stated that she felt that the treatment of accidentals was too irregular to make the manuscript useful for research in that field. This is certainly the case when looking at single pieces, and it is very difficult to spot patterns over the whole manuscript through traditional analysis. However, techniques in computer-aided symbolic music analysis now allow us to search for patterns in large datasets, in this case the Buxheim Organ Book and almost all of its concordances, encoded in Humdrum format. In my previous research on cadential accidentals (Acta Musicologica, 2011), clear patterns of notation emerged, allowing me to reconstruct the decisions made by the main scribe of the Buxheim manuscript, contrasted against the background of its concordances. In this article I apply a similar method to the analysis of signature accidentals: notated in Old German organ tablature, Buxheim has no signatures; accordingly, all accidentals must be explicitly notated. Through computer-aided analysis, I found patterns in how signature accidentals from sources in mensural notation are translated into organ tablature. These patterns inform our understanding of the prescriptive nature (or not) of partial signatures, cancellation of signature accidentals for cadential and other purposes, transposition, and the scribe’s possible attitude to modal identity and classification.

A quantitative study of seven historically informed performances of Bachs BWV1007 Prelude In the field of early music, the urge to realize historically informed interpretations has led to new perspectives about our musical legacy from scholars and performers alike. Consequently, different schools of early music performance practice have been developed through the 20th and 21st centuries. Analysing the development of performance practice and aesthetics from the perspective of cognitive and computational musicology can potentially provide insights into interrelations between musicology, performance practice and cognition. In the field of performance analysis we may therefore aim to find out which aspects are involved in the categorization of performances or in the identification and understanding of possible performance trends that change over time. For instance, are sudden changes in tempo within the performance of a piece representative of certain musicologically informed interpretations? Is this approach communicated by the performance itself? Do perception and cognition play a role in the aesthetic choices involved in performance? This article concentrates on applying three state-of-the-art quantitative methodologies in expressive performance analysis to elucidate possible relations among musicological and cognitive interpretations. It is shown how a particular methodology may serve (or constrain) the ability to compare different interpretations and define ‘expressiveness’. For this purpose, an analysis and comparison is presented of seven performances of the Prelude from J. S. Bach’s Cello Suite no.1 in G major, bwv1007, played by three different performers.

Reidentifying the wood of the Queen Mary and Lamont harps Identifying the wood of the surviving historical wire-strung harps of Ireland and Highland Scotland has long been an important goal of researchers and instrument-builders. In 1969, microscopic examination of the anatomical features of the wood of two of the earliest surviving harps of this type, the Queen Mary and Lamont of National Museums Scotland, identified all parts of both as European hornbeam (Carpinus betulus). Due to the importance of these two harps as early exemplars, this identification has had far-reaching implications for understanding the construction practices for this type of historical harp. Questions about the identification of the woods of the Queen Mary and Lamont harps have prompted a re-evaluation. In this article, we discuss the observations and evidence that led to our decision to reidentify the wood of both harps, including the use of X-ray computed tomography to test the earlier identification of all wooden members as the same species. A rigorous new identification has been undertaken that addresses the fragile state of the wood and the need to minimize the impact on these important musical artefacts by revising the sampling method and utilizing scanning electron microscopy as an alternative to conventional microscopic examination. The results of our work to date are presented, and the implications for these two harps and for other harps of this type are discussed.

Exploring information retrieval, semantic technologies and workflows for music scholarship: the Transforming Musicology project Transforming Musicology is a three-year project undertaking musicological research exploring state-of-the-art computational methods in the areas of early modern vocal and instrumental music (mostly for lute), Wagner’s use of leitmotifs, and music as represented in the social media. An essential component of the work involves devising a semantic infrastructure which allows research data, results and methods to be published in a form that enables others to incorporate the research into their own discourse. This includes ways of capturing the processes of musicology in the form of ‘workflows’; in principle, these allow the processes to be repeated systematically using improved data, or on newly discovered sources as they emerge. A large part of the effort of Transforming Musicology (as with any digital research) is concerned with data preparation, which in the early music case described here means dealing with the outputs of optical music recognition software, which inevitably contain errors. This report describes in outline the process of correction and some of the web-based software which has been designed to make this as easy as possible for the musicologist.

Writing a Big Data history of music This article introduces the project A Big Data History of Music, which set out to unlock the bibliographical data held by research libraries in order to create new research opportunities for musicologists. The project cleaned and enhanced aspects of the British Library catalogues of printed and manuscript music, which are now available as open data. It also experimented with the analysis and visualization of the British Library datasets and the RISM inventories of printed and manuscript music. The article shows how quantitative analysis of these datasets can expose long-term historical trends, such as the rise and fall of music printing in 16th- and 17th-century Europe. Data analysis and visualization also facilitates research on the dissemination and canonization of specific composers (as shown by case-studies on Palestrina and Purcell) and on changing trends in genres, scoring and ethnic colourings in music (as shown by a case-study on ‘Scottish’ music).

Text mining and early music: using Lexomics in research Amongst the digital research methods that have developed in recent years, one with immense possibilities for early music scholarship is text mining. This approach extracts data from written sources, determining patterns based on statistical analyses of word usage. Because this approach is computer-based, it can quickly process a large number of sources at a single time. While text mining cannot replace other traditional research methods, a program such as Lexomics (http://wheatoncollege.edu/lexomics/introduction-lexomics/) is a useful addition to early music scholarship. Lexomics is a tool that determines relationships between text sources based on an analysis of word frequency. It can reveal connections or disparities between or within sources with regard to vocabulary, style, potential authorship and transmission. When applied to early theoretical treatises, it can therefore illuminate relationships between greater- and lesser-known sources or authors, trace textual influences, expose shifts in terminology or assist in the re-creation of potential stemmas. In this article, I provide an overview of the Lexomics program and demonstrate its practical usefulness in early music research. Using texts from the online Thesaurus Musicarum Latinarum database (and the critical editions from which they were drawn), I apply Lexomics to one of my own current projects: an exploration of the relationship between Petrus de Sancto Dionysio’s Tractatus de musica and a set of small anonymous treatises that circulate alongside in its manuscript sources.