Journal of English Linguistics

This essay introduces a special issue exploring strategies for complicating corpus methods in historical English language studies: both the ways that corpus methods can accommodate the complications of historical work and the ways that scholars can productively complicate the tools in order to enrich our use of digital archives. It highlights limitations of data, issues of variable orthography, and deductive versus inductive approaches. The essay also addresses why it is valuable to use mixed methods, statistical models, LLMs, and interdisciplinary approaches in corpus work—insights that may benefit scholars working with language from any historical period, including the present day. This introduction provides brief summaries of the three research articles in the special issue.

Electronically available corpora have inevitably enhanced access to historical texts, but building historical corpora can often, unintentionally, distort historical language. This paper discusses how issues of faithfulness and sampling can impact the reliability of a historical corpus-based analysis, with recommendations provided for both corpus users and corpus builders. First, by comparing the Beowulf manuscript with the language found in a historical corpus, this paper shows how seemingly innocent editorial changes such as spacing, punctuation, and spelling can have linguistic consequences for phonological, morphological, syntactic, and socio-historical analyses. The removal or addition of elements is argued to have both theoretical and methodological implications. Second, this paper discusses how sampling issues can lead to erroneous conclusions by corpus users about historical language. Recommendations are provided regarding supplementation and mixed-method approaches.

Turning text into numbers—and ultimately into an interpretable result—typically involves a processing pipeline. Whether explicit or hidden inside code, each step in that pipeline involves a host of decisions. This article follows a pipeline and unpacks some of the attendant decisions that analysts face at various stages, highlighting the specific challenges that confront analysts investigating historical data. Data, for example, can be less available; the quality of those data can be poor; and the development of tools and models can lag when compared to what is available for present-day English. In the face of such challenges, options are explored, but no argument is made for a particular approach or method—results and their interpretation will be shaped by analysts’ epistemic commitments and expertise. An argument is made, however, that an effective processing pipeline will be generative. It will be one that not only can be explained and defended but also can help us see the histories of English—its structures, its uses, and its users—in new ways. In that spirit, a processing pipeline can be understood not as a series of immobilizing decisions, but as an invitation.

This article analyzes the frequency of three clause-level coordinating conjunctions, and, but, and or, in addition to a wide range of subordinating conjunctions across the Transhistorical Corpus of Written English (TCWE). This resource includes five genres covering seven centuries: sermons and statutes dating from the fifteenth to twenty first centuries, personal letters dating from the fifteenth to twentieth centuries, and emails and WhatsApp messages dating from the twenty-first century. The study engages with three key developments in the history of English: the increase in the inventory of subordinators between 1500 and 1700, the development of expository and non-expository registers in English between 1650 and 1990, and the increase in colloquialization moving into the Present-Day English period. It finds that there is more substantial genre-conditioned variation in relation to all three developments than has previously been acknowledged. Its results also contribute to knowledge about shifts in the generic conventions of English sermons and statutes over time, and language use within email and instant messaging, most notably a preference for causal as over because in the instant messaging sample.