Simon Fraser University
Datamining tools such as Naïve Bayes, which seek to
extract and quantify textual features, do not appear
to support pedagogies, research practices, or facilitate
the pleasures of discovery in the same way humanist researchers
have become used to (Rommel 2004, Harley et
al. 2006). Jean Guy Meunier, Ismail Biskri, and Dominic
Forrest (2005) argue that reading or analyzing text is “a
more complex procedure of heuristics and pattern recognition
strategies. All of which are grounded on complex
dimensions such as linguistic and mundane semantic
structure, inference, pragmatic memory, culture, social
interaction, and knowledge repositories” (124, 126).
They conclude the “text does not necessarily reveal itself
in a first reading, not even in a first analysis” and that
the “computer can play a productive role in the reading
and analysis process; but only if it is well-situated as an
assistant to such cognitive activities” (126). Although in
its current stage, the MONK Workbench may be most
helpful to domain experts who are already familiar with
the social, legal, or literary phenomenon they study, it
can be used to facilitate pattern finding across corpora
large enough to loosen the threads that tie texts together.
My original research question asked if analytics used
within the MONK Workbench could help locate Richard
Head’s biography of Mother Shipton, not within
the fifteenth-century, when he claims it was originally
written, but within the late sixteenth century when it was
published. While reporting on these preliminary findings,
this paper will likewise suggest that, although the
tools can render texts into composite features, they can
also suggest ways in which texts can be meaningfully realigned.
The serendipitous associations created by these
re-alignments support an engagement with texts that is
familiar to research scholars working outside of digital
humanist practices.
Mother Shipton and the Devil
Mother Shipton may have lived in the late 15th to mid
16th centuries (1488-1561?); she emerged into the literary
record in 1641 as an early modern prophetic witch
whose embodied spirituality was made her spiritual
allegiances hard to identify. Her prophecies were ultimately
created or appropriated by authors who tantalizingly
propose that her predictive skill and bodily deformities
illustrated an otherworldly or demonic heritage.
The inclusion of the Gentleman Devil, a sucked witch’s
mark, and animal familiars in Richard Head’s biographies
of Mother Shipton situate these texts alongside the
prophetic norms and evolution of theories of witchcraft
and diabolism written during the seventeenth century. In
hoping to exploit interest in ecstatic prophecy, practical
maleficium, and, most crucially, the Devil’s relationship
to the witch, Head’s biographies conform too closely to
seventeenth-century stylistic elements, ultimately revealing
their fictions. The devil is in the details.
I theorize that the presence of this seducing gentleman
devil could help situate Head’s construction of Mother
Shipton within a context of continental beliefs about
witchcraft emerging in seventeenth-century trial accounts.
In England, the witch’s mark appeared most often
as a scratch, or a prick, then as a teat from which
the witch’s familiar was supposed to drink as reward for
the malefic crimes it committed at the witch’s behest.
References to having sex with the devil became prominent
in England as a reflection of newly imported continental
beliefs about witchcraft. The devil in the shape
of a gentleman, or the “black man,” who asks to suck
from these increasingly sexualized marks, appears after
the sixteenth-century when Head’s biography of Shitpon
was supposedly written. At the beginning of my MONK
research I created a paper dataset with pens and highlighters.
With this research, I found a tract, “The Mystery
of Witchcraft” (1626), which showed the presence
of “carnall knowledge of [the witches] body” earlier than
I’d anticipated. In finding this, I had to rethink when the
idea of demonic sex entered the understanding of English
witchcraft, but also whether coitus was the only demonic
sex act present in the material.
The MONK Workbench combines “texts and tools to enable
literary research through the discovery, exploration,
and visualization of patterns. . . . Each toolset is made up
of individual tools (e.g. a search tool, a browsing tool,
a rating tool, and a visualization), and these tools are
applied to worksets of texts selected by the user from
the MONK datastore” (MONK, Workbench). I rated a
large sample set of fifty-two texts using the Search by
Example tool, and rated them as containing “sex,” “no
sex,” and “some sex” with the devil, and ran a Naïve
Bayes analytical routine, asking for lemma as my feature,
and requesting one hundred “features” (words that
are representative of each class of documents) be returned.
Some interesting results emerged. For example,
the later witchcraft tract, A True and impartial relation of
the informations against three witches, viz., Temperance
Lloyd, Mary Trembles, and Susanna Edwards, who were
indicted, arraigned and convicted at the assizes holden
for the county of Devon, at the castle of Exon, Aug. 14, 1682 was returned as a match for “sex.” Here is a suggestive
excerpt found in the tract:
The said Informant upon his Oath saith, That upon the
17th day of July instant this Informant did hear Susanna
Edwards to confess, that the Devil had carnal knowledge
of her Body; and that he had suckt her in her Breast and
in her Secret parts.
The Workbench analytics returned From Newes from
Scotland, Declaring the Damnable life and death of
Doctor Fian, a notable Sorcerer, who was burned at
Edenbrough in Ianuary last. 1591 as a positive hit for
“sex” with the devil. This excerpt, extracted from a large
enough text chunk to make it difficult to locate by eye,
comments, perhaps unintentionally, on the more pleasurable
kinds of malefic sexuality seen in these encounters:
for as much as by due examination of witchcraft and
witches in Scotland, it hath latelye beene found that the
Deuill dooth generallye marke them with a priuie marke,
by reason the Witches haue confessed themselues, that
the Diuell dooth lick them with his tung in some priuy
part of their bodie, before hee dooth receiue them to
be his seruants, which marke commonly is giuen them
vnder the haire in some part of their bodye, wherby it
may not easily be found out or seene, although they be
searched
Although these texts do not share significant linguistic
features, and the second excerpt is not in fact a story of
English witchcraft, but of Scottish, this represents a very
positive result in terms of finding patterns of malefic sexuality
across a corpus. Although they suggest the necessity
of rethinking how sexuality functions in witchcraft
tracts and as such proves to be a useful to look at familiar
texts, their location within large text chunks would have
obscured them to non-domain experts.
The most provocative results of the sample was a serendipitous
association that contextualized Richard Head’s
construction of Mother Shipton’s diabolical upbringing.
The MONK Workbench provided an opportunity
to “search by example” and find unpredicted associations
in terms of this research stream. The Workbench
analytics rated a text as “sex” with the devil, which I
had not considered, a text about Tannakin Skinker, the
“hog-faced woman” from Holland. Head’s text and the
Skinker text share themes such as implied witchcraft, the
presence of the Devil, and the implication of the Devil’s
presence, the presence of a witch or a witch’s curse, the
use of wealth as a lure, the use of costly clothing as a tool
for seduction, the presence of a castle or a large home,
and, most substantially, hog-faced children. These numerous
alignments are suggestive of the kinds of source
text or framework upon which Head drew an expanded
version of Shipton’s life. Although malefic sex was identified
as a signifier of diabolism, numerous texts rated as
“sex” with the devil by the Workbench analytic process
included prodigious birth as a sign of the Devil’s presence.
Prodigious births, along with witchcraft, comets,
and two-headed chickens, belongs to the same kind of
tabloid, or “enquirer,” genre in early modern England;
Head’s version of Shipton’s life falls firmly within that
genre in a way which makes the devil’s presence in it
make sense outside of witchcraft or prophecy.
Conclusions
In this way, the computational process assisted my reading
process rather than delineated it as Meuneir et al. argue
must happen to make computational analytics more
useful to main stream critical inquiry. Although Head’s
version of Shipton’s life is not a text in the dataset, I
was able to use the MONK Workbench to find a set of
texts within the same genre. That is, I was able to find
comparable texts without a base text against which they
were compared. In addition, while the “predicted features”
were only moderately helpful, patterns began to
emerge that reified and challenged my understanding of
the devil’s evolution in sixteenth-century witchcraft accounts
and therefore my understanding of Mother Shipton’s
location within these texts. Although preliminary
trials with Workbench analytics produced mixed results,
only sometimes predicting that texts contained the elements
I was looking for to help contextualize Richard
Head’s version of Mother Shipton within sixteenth century
ephemera, it suggested connections based on similarities
and patterns I had not predicted. As a result, I
reevaluated my hypothesis regarding demonic sexuality
and its role in the prodigious birth category of early
English ephemeral publications. Ultimately, this experience
facilitated serendipitous literary discoveries based
on unpredictable features, a process of discovery that
makes tools such as those incorporated in the MONK
Workbench more user-friendly to users unfamiliar with
current digital research methods.
Works cited
Anon. A True and impartial relation of the informations
against three witches, viz., Temperance Lloyd, Mary
Trembles, and Susanna Edwards, who were indicted, arraigned
and convicted at the assizes holden for the county
of Devon, at the castle of Exon, Aug. 14, 1682
Anon. The most strange and admirable discouerie of the
three witches of Warboys. London: Printed for Thomas
Man and Iohn Winnington, and are to be solde in Pater
noster Row, at the signe of the Talbot. 1593. Anon. The prophesie of Mother Shipton in the raigne
of King Henry the Eighth. London: Printed for Richard
Lownds, at his Shop adjoyning to Ludgate, 1641.
Anon. The Strange and wonderful history of Mother
Shipton. London: Printed for W.H. and sold by J. Conyers,
1667/1668?
Head, Richard. The life and death of Mother Shipton,
London : Printed for B. Harris 1677.
Harley, Diane, et al. Use and Users of Digital Resources:
A Focus on Undergraduate Education in the Humanities
and Social Sciences (2006) Center for Studies in Higher
Education, University of California, Berkeley. 11-06
Meunier, Jean Guy, Ismail Briski, and Dominic Forest.
“A Model for Computer Analysis and Reading of Text
(CARAT): The SATIM Approach” Text Technology,
Number 2, 2005. 123-151
MONK Project. “Introduction” Monk Workbench. Online.
Rommel, Thomas. “Literary Studies.” A Companion to
Digital Humanities.
SCHREIBMAN, SUSAN, RAY SIEMENS and JOHN
UNSWORTH (eds). Blackwell Publishing, 2004. Blackwell
Reference Online. 22 September 2008 http://www.
blackwellreference.com.login.ezproxy.library.ualberta.
ca/subscriber/tocnode?id=g9781405103213_chunk_
g978140510321311
If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.
Complete
Hosted at University of Maryland, College Park
College Park, Maryland, United States
June 20, 2009 - June 25, 2009
176 works by 303 authors indexed
Conference website: http://web.archive.org/web/20130307234434/http://mith.umd.edu/dh09/
Series: ADHO (4)
Organizers: ADHO