Glossary
A list of technical terms used in various posts, defined and linked to posts where they were used.
- Ancestral text
-
A name for the reconstructed or idealized text that stands behind surviving witnesses in editorial and computational work. The post that introduces it asks what happens when digital methods treat that ancestor as a measurable object rather than only a scholarly fiction.
See The Ancestral Text .
- Biplot
-
A scatterplot that shows both the texts under study and the linguistic features that pull them apart, usually derived from a principal component analysis. On this site it is a standard way of reading a corpus at a glance.
See A Map of Early English Print , Finding “Distances” Between Shakespeare’s Plays 1 , Finding “Distances” Between Shakespeare’s Plays 2: Projecting Distances onto New Bases with PCA .
- Corpus
-
A collection of texts assembled for comparison, with enough shared preparation that counts across items remain meaningful. Much of the later work here depends on corpora built for Visualizing English Print and on the transcribed books of EEBO-TCP.
See Visualizing English Print, 1530 -1800: The Corpus, Tag Sets, and Topics , Visualizing English Print, 1530-1800, Genre Contents of the Corpus , The Great Work Begins: EEBO-TCP in the wild .
- Critical gesture
-
A basic move of interpretation, such as pointing to a passage, circling a pattern, or naming a structure, treated here as part of an ecology of critical practice. The phrase is meant to make ordinary scholarly habits available for comparison with computational ones.
See An Ecology of Critical Gestures: Point, Circle and Name .
- Dendrogram
-
A tree diagram produced by hierarchical clustering, showing which texts group together when measured by linguistic similarity. Early posts use dendrograms to sort Shakespeare’s plays and other early modern drama before the larger VEP maps appear.
See Shakespearean Dendrograms , Clustering the Plays Without Principal Components , A Genre Map of Shakespeare’s Plays from the First Folio (1623) .
- Distance
-
A numerical measure of how far apart two texts stand when each is represented by a profile of linguistic features. Several posts ask what such distances mean for genre and for the internal shape of Shakespeare’s dramatic career.
See Finding “Distances” Between Shakespeare’s Plays 1 , Finding “Distances” Between Shakespeare’s Plays 2: Projecting Distances onto New Bases with PCA .
- DocuScope
-
A text-analysis system that tags English prose and verse for rhetorical and functional categories, then returns counts that can be compared across documents. It is the principal tagging engine behind the Shakespeare genre maps and much of the later corpus work with Jonathan Hope.
See A Genre Map of Shakespeare’s Plays from the First Folio (1623) , Comic Twelfth Night, Tragic Othello? (Part I) , Shakespeare Quarterly Article Goes Live , A Map of Early English Print .
- EEBO-TCP
-
Early English Books Online, Text Creation Partnership: a large set of transcribed early English printed books, released in phases for scholarly use. Jonathan Hope’s post on its arrival treats the shift from working with dozens of texts to tens of thousands.
See The Great Work Begins: EEBO-TCP in the wild , A Map of Early English Print .
- Feature
-
A countable linguistic category used to represent a text as a vector of frequencies, for example a DocuScope tag or a part-of-speech pattern. Features are what the statistical procedures on this site actually compare.
See A Genre Map of Shakespeare’s Plays from the First Folio (1623) , Adding Knobs to the Analysis , Visualizing English Print, 1530 -1800: The Corpus, Tag Sets, and Topics .
- First Folio
-
The 1623 collected edition of Shakespeare’s plays, often used here as a closed set for early experiments in genre and clustering. Later work widens the frame from that collection to early modern drama and print more generally.
See A Genre Map of Shakespeare’s Plays from the First Folio (1623) , Shakespearean Dendrograms .
- Fuzzy structuralism
-
A name for critical work that looks for patterned relations in language without insisting on hard boundaries between categories. It describes a stance taken in several posts toward genre, influence, and the results of clustering.
- Genre
-
On this site, not only a traditional literary label but a pattern that can be recovered from linguistic counts across a set of plays or books. Early posts ask whether comedy, tragedy, and history leave distinct signatures in Shakespeare’s language.
See A Genre Map of Shakespeare’s Plays from the First Folio (1623) , Comic Twelfth Night, Tragic Othello? (Part I) , Comic Twelfth Night, Tragic Othello (Part 2) , Comic Twelfth Night, Tragic Othello (Part III) .
- Ideolect
-
The distinctive linguistic habits of a speaker or character, treated here as something that may shape the genre profile of a play. Student work on the blog asks how far comedy or tragedy depends on such local habits of speech.
See Genre Dependence on Character Ideolects? (by Mike Stumpf, UW Undergrad) .
- Invisible grammar
-
Patterns of function words and syntactic habit that readers rarely notice but that computational counts can recover. Paired in one post with Shakespeare’s more conspicuous “mythic vocabulary,” it names the quiet regularities that organize dramatic language.
See Shakespeare’s mythic vocabulary – and his invisible grammar .
- LATtice
-
A visualization tool used here to display linguistic variation across many texts at once, making local and global patterns easier to inspect. It appears in the middle years of the blog as a way of seeing variation without collapsing everything into a single plot.
- Linguistic variation
-
Systematic difference in language across texts, genres, authors, or periods, treated as something that can be measured rather than only described. It is the shared object of Witmore and Hope’s collaboration across more than a decade of posts.
See A Genre Map of Shakespeare’s Plays from the First Folio (1623) , Visualizing Linguistic Variation with LATtice , Digital approaches to the language of Shakespearean Tragedy , A Map of Early English Print .
- Local and diffused variation
-
A contrast between differences concentrated in a short stretch of text and differences spread through a whole work. One early post uses the Hinman Collator as an analogy for seeing both kinds of variation in Shakespeare.
- Massively addressable object
-
A way of describing a text once digital methods make it possible to address the same work at many scales, from the word to the corpus. The phrase names a shift in what kind of object a literary text can be for criticism.
See Text: A Massively Addressable Object , Texts as Probability Clouds .
- Metadata
-
Structured information about texts in a corpus, such as date, genre, author, or format, kept separately from the words themselves. Posts on Visualizing English Print treat clean metadata as a condition of large-scale comparison.
See Data and Metadata , Visualizing English Print, 1530 -1800: The Corpus, Tag Sets, and Topics .
- n-gram
-
A contiguous sequence of n words or characters used as a unit of comparison, from single words (1-grams) to short phrases. Posts here use trigrams and Google n-gram data to test authorship problems and questions of philosophical vocabulary.
See Edward III, Shakespearean Trigrams, and Trillin’s Derivatives , Google n-grams and Philosophy: Use Versus Mention , Lost Books, “Missing Matter,” and the Google 1-Gram Corpus .
- Outlier
-
A text that sits apart from its expected group when plays or books are clustered by linguistic features. Early Shakespeare posts treat outliers as prompts for critical attention rather than as errors to be discarded.
- Phylogenetic inference
-
Methods borrowed from evolutionary biology that build trees of relationship from shared traits, here applied to linguistic features of texts. Jonathan Hope explores what such trees can and cannot claim about literary history.
See Phylogenetic inference .
- Pointing game
-
A description of work with large language models that treats the exchange as shared attention to passages and problems rather than as conversation with a mind. It is the governing metaphor of the archive’s last post.
See AI and the Pointing Game .
- Principal component analysis
-
A statistical method that reduces many correlated feature counts to a few axes that capture most of the variation in a corpus. On this site PCA is the usual route from DocuScope counts to maps and biplots of early English print.
See Finding “Distances” Between Shakespeare’s Plays 1 , Finding “Distances” Between Shakespeare’s Plays 2: Projecting Distances onto New Bases with PCA , A Map of Early English Print , Mapping the ‘Whole’ of Early Modern Drama .
- Probability cloud
-
A figure for the text once it is understood as a distribution of possible addresses and counts rather than as a single fixed object. It belongs with the argument that digital methods change what kind of thing a literary work is.
See Texts as Probability Clouds , Text: A Massively Addressable Object .
- Rigid classifier
-
A sorting rule that assigns texts to categories by fixed criteria and then struggles when historical change moves the boundary. Work with Robin Valenza uses the idea to describe problems of dating and periodization in large corpora.
See The Time Problem: Rigid Classifiers, Classifier Postmarks .
- Speech prefixes
-
The abbreviated speaker names that introduce lines in a printed play. Preparing drama for analysis often means deciding how to handle these labels so that they do not distort counts of the spoken text.
See Data and Metadata , Visualizing English Print, 1530 -1800: The Corpus, Tag Sets, and Topics .
- Tag set
-
The inventory of categories a tagging system such as DocuScope can assign, together with the rules that map words and phrases onto those categories. Different tag sets yield different pictures of the same corpus.
See Visualizing English Print, 1530 -1800: The Corpus, Tag Sets, and Topics , Adding Knobs to the Analysis .
- Texts as objects
-
An early line of inquiry that asks what follows if literary works are treated as objects among other objects, with properties that can be counted and compared. The posts bring object-oriented philosophy into contact with digital criticism.
See Texts as Objects I: Object Oriented Philosophy. And Criticism? , Texts as Objects II: Object Oriented Philosophy. And Criticism? .
- Topic modeling
-
A family of unsupervised methods that infer recurring clusters of words across a large collection of documents. It appears here especially in work with Eric Alexander and in the documentation of the VEP corpora.
See Visualizing English Print, 1530 -1800: The Corpus, Tag Sets, and Topics , Digital approaches to the language of Shakespearean Tragedy .
- Visualizing English Print
-
A Mellon-supported project based at Wisconsin (with collaborators elsewhere) that built corpora, metadata, and tools for studying English printed books from the sixteenth century through 1800. Many later posts on this site report or extend that work.
See Visualizing English Print, 1530 -1800: The Corpus, Tag Sets, and Topics , Visualizing English Print, 1530-1800, Genre Contents of the Corpus , Data and Metadata , A Map of Early English Print .
- Ward clustering
-
A hierarchical clustering method that groups items so as to minimize the increase in within-group variance at each step. Several early Shakespeare experiments on the blog use Ward clustering to produce dendrograms of the plays.
See Shakespearean Dendrograms , Clustering the Plays Without Principal Components .
- WordHoard
-
A tool for searching and comparing lemmatized early modern texts, used in posts that examine word frequency and the language of individual plays. It appears especially in work on Macbeth and related Shakespearean vocabulary.