Complementary critical traditions and Elizabeth Cary's Tragedy of Mariam

It has become a critical commonplace that computational
studies have not achieved widespread acceptance
in mainstream literary studies (Corns 1991, Potter
1991, Fortier 1991, McGann 2001, Ramsay 2003, Rommel
2004). It is also the case, however, that practitioners
have tended to be cautious about the sorts of interpretive
claims that can be made on the basis of statistical analysis.
Ramsay (2003) locates the source of the caution in
a from of “dissonance” in the computing community
where empirical studies are seen as providing a kind of
safe haven that would be compromised if the researcher
were to intervene excessively with either the data or the
analysis (168).
The view of computational stylistics as “scientific” and
therefore irrelevant to mainstream literary criticism is
a pervasive one. Scholars who are sympathetic to the
methods involved in attribution and the analysis of style
typically construct the problem of humanities computing
relationship with the wider scholarly community in
terms of a need for computing scholars to rethink their
research such that it is framed in terms of “interesting”
versus “right” (Ramsay 2003 173) or “interactive tool”
versus “quantitative tool” (Sinclair 2003 177) or “disciplined
play” versus “unity and coherence” (Rockwell
2003 213). In constructing the debate in these terms there
is a risk of amplifying whatever differences there may be
between traditional humanities scholars and computing
scholars. The result is that the perceived differences become
further entrenched. It also sidesteps the question of
how and why attribution studies work.
Attitudes to what can be said about a text depend largely
on the account of language that underpins an analysis
(Connors 2006a; Hoover 2007). Cognitive grammar as
developed by Ronald Langacker (1987, 1991) provides
a theory that justifies the counting of function words in a
computational analysis and also provides an interpretive
framework that can explain the use of function words
as a rhetorically motivated choice with semantic implications.
As such it provides a promising theoretical account
for the kinds of computational studies carried out by Burrows (1987a, 1987b, 1992b), Craig (1999a, 2002,
2004), Hoover (2003a, 2003b), Rybicki (2005), and others.
This study explores the use of function words in an analysis
of style of 60 tragedies published between 1580 and
1641, with a particular focus on Elizabeth Cary’s, The
Tragedy of Mariam. The set is comprised of the 12 socalled
“Sidnean” closet tragedies, as well as 48 tragedies
that were written for performance (See Appendix 1).
Of these 5 were performed in private theatres and 41 in
public theatres. It is not known whether the remaining 2
were performed (Harbage & Schoenbaum, 1964). Some
textual preparation was carried out prior to the analysis,
but editing and coding was kept to a minimum. Although
spelling was regulated homographs were not tagged.
Contracted forms throughout the texts were expanded so
that their constituents appeared as separate words.
Tagged texts were then run through a frequency count
using a program developed by the Centre for Literary
and Linguistic Computing (CLLC) at the University of
Newcastle called Intelligent Archive (IA). A list of 241
function words was used (see Appendix 2). This list was
developed on the basis of work by the CLLC. Using IA a
frequency test was run to establish which function words
were most commonly found in each of the plays. The
frequency count for each word in the function word list
was expressed in the IA output as a percentage of the
total dialogue in the relevant play. The data produced by
IA were then transferred to Statistical Package for the
Social Sciences (SPSS), and Excel, for further analysis. On the basis of strong discriminate analysis results an
independent samples t test was carried out to identify a
set of “closet” and “stage” markers (see appendix 3). The test for equality of means and Mann Whitney test. A
Principal Component Analysis (PCA) was then carried
out to identify which variables were responsible for most
of the difference between the sets of texts. For the PCA,
the texts were broken into 4000 word segments to ensure
that results could be easily presented. Segments of less
than 4000 words were incorporated into the preceding
segment. This produced 268 segments in total (50 closet
segments and 218 stage play segments). The results of
the PCA are set out in Table 1 and Table 2. One of the most interesting features of Figure 2 is the
way the graph identifies three loosely clustered generic
groupings. These are closet plays to the east, public stage
plays to the west, and private stage plays along with
Mariam in the centre. Marta Straznicky (2004) analyses
Mariam in the context of drama associated with the private
theatres. She notes that writers for the stage actively
incorporated elements of the neo-Senecan tradition associated
with closet tragedy into their plays. While some
writers for the stage adopt various features commonly
found in closet tragedy to a greater extent than their colleagues,
Elizabeth Cary appears to draw on a more theatrical
tradition than her closet contemporaries.
This is an interesting result for Elizabeth Cary’s play because
it supports the insights of more traditional humanities
scholars who have described Mariam as the most
“theatrical” of the Sidnean closet tragedies (Barish 1993;
Findlay, Williams & Hodgson-Wright 1999; Straznicky
2004). The play features sub-plots, stage directions and
action that is played out rather than dealt with through
narrative. Function word analysis supports the view of
Mariam as one of the more theatrical of the closet texts,
and provides additional information about how Mariam
relates to contemporary texts in terms of the selected variables. One of the strongest differences between the plays written
for the stage, and closet plays, is the use of pronouns.
There is evidence of frequent usage of the first and second
person singular pronouns in the stage plays and little
evidence of their use in closet plays. Van Hoek (2007)
has analysed pronouns from the perspective of Langacker’s
cognitive grammar, particularly in relation to the
way they influence the accessibility of concepts and the
viewpoint from which constructions are understood. Van
Hoek’s analysis can usefully explain aspects of closet
tragedy and The Tragedy of Mariam which have caused
readers over time to find the text(s) less engaging than
plays written for the public stage.
In a cognitive analysis, language that features frequent
use of pronouns, particularly anaphoric pronouns, (such
as him in “I’ll go and get him”) assumes a certain common
ground between the speaker and the addressee; they
both understand who or what is being referred to without
“him” being referred to by name. When a full-noun is
used, like a name, there is an implied conceptual distance.
The notion of “conceptual distance” also relates
to how easily accessible a concept is for the discourse
participants.
In her first soliloquy, a speech of seventy-five lines, Mariam
refers to herself by name four times. She asks herself
“then why grieves Mariam Herod’s death to hear?”
(I.i.38) and calls herself “heard-hearted Mariam” (I.i.62).
In most situations it is unusual for a speaker to refer by
name to himself or herself. When Mariam names herself,
she places the conception of “Mariam” on the metaphorical
stage shared between speaker and addressee.
In cognitive linguistic terms, it is as though Mariam is
observing herself from an outside position. It’s a device
that breaks up the normal relationship between speaker
and addressee and has a disorienting effect.
In his analysis of the Senecan tradition in England,
Braden asks “what makes it appropriate for a character
to talk this way?” (1985 66). There could be a number
of reasons for Cary’s characters to engage in this type of
self-referential linguistic behaviour, however it is likely
that the style of speech is rhetorically motivated. Although
the language of the play is somewhat alienating,
it reflects serious philosophical, cultural and political
ambitions. Mariam is explicitly concerned with the individual’s
right to resist tyranny. The form this resistance
takes reflects the influence of neo-Stoic discourses and
Seneca as literary and philosophical model.
For Braden the characteristic feature of Senecan drama
is a kind of “cosmic self-dramatization” (178) that is
represented in the characters by a quest for “a radical,
unpredicated independence” (67). The speech of Cary’s
characters can be linked to this very neo-Senecan idea of
selfhood and the neo-Stoic ideals of control of the self,
especially in the face of unchecked abuse of power. Mariam’s
conquest of self can be traced through the way she
overcomes emotional confusion to become “free, lofty,
fearless and steadfast” (Seneca qtd Straznicky 1994
115). Evidence of this struggle in Mariam can be seen
at multiple levels, including the linguistically at the level
of function words.
References
Barish, Jonas. “Language for the Study: Language for
the Stage.” The Elizabethan theatre XII. Eds. A. L. Magnusson
and C. E. McGee. Toronto: P.D. Meany, 1993.
19-43.
Braden, Gordon. Renaissance Tragedy and the Senecan
Tradition: Anger’s Privilege. New Haven: Yale University
Press, 1985.
Burrows, J.F. Computation into Criticism: A Study of
Jane Austen’s Novels and an Experiment in Method.
New York: Oxford University Press, 1987a.
---. “Computers and the Study of Literature.” Computers
and Written Texts. Ed. C.S. Butler. Oxford: Blackwell,
1992b.
---. “Word Patterns and Story Shapes: The Statistical
Analysis of Narrative Style.” Literary and Linguistic
Computing 2 (1987b): 61-70.
Cary, Elizabeth (Lady Falkland). The Tragedy of Mariam
the Faire Queene of Jewry. London: Richard Hawkins,
1613.
Connors, Louisa. “Linking Cognitive Linguistics and
Computational Stylistics.” ALLC Digital Humanities.
Sorbonne Paris IV: Centre Culture Anglophones et Technologies
de l’Information (CATI), 2006a. 46-47.
Corns, Thomas N. “Computers in the Humanities: Methods
and Applications in the Study of English Literature.”
Literary and Linguistic Computing 6.127-30 (1991).
Craig, D.H. “Common-Word Frequencies, Shakespeare’s
Style, and the Elegy by W.S.” Early Modern
Literary Studies 8.1 (2002): 1-42.
Craig, Hugh. “Authorial Attribution and Computational
Stylistics: If You Can Tell Authors Apart, Have You Learned Anything About Them.” Literary and Linguistic
Computing 14.1 (1999a): 103-12.
---. “Stylistic Analysis and Authorship Studies.” A Companion
to Digital Humanities. Eds. Susan Schreibman,
Ray Siemens and John Unsworth: Blackwell, 2004. 273-
88.
Findlay, Alison, Gweno Williams, and Stephanie Hodgson-
Wright. “”The Play Is Ready to Be Acted”: Women
and Dramatic Production, 1570-1670.” Women’s Writing
6.1 (1999): 129-48.
Fortier, P.A. “Theory, Methods and Applications: Some
Examples in French Literature.” Literary and Linguistic
Computing 6.192-6 (1991).
Harbage, Alfred, and S. Schoenbaum, eds. Annals of
English Drama 975-1700: An Analytical Record of All
Plays, Extant or Lost, Chronologically Arranged. 1940.
Second ed. Philadelphia: University of Philadelphia
Press, 1964.
Hoover, D.L. “Another Perspective on Vocabulary Richness.”
Computers and the Humanities 37 (2003b): 151-
78.
---. “The End of the Irrelevant Text.” Digital Humanities
Quarterly 1.2 (2007).
Hoover, David L. “Multivariate Analysis and the Study
of Style Variation.” Literary and Linguistic Computing
18.4 (2003a): 341-60.
Langacker, Ronald W. Foundations of Cognitive Grammar:
Descriptive Application. Vol. II. Stanford, California:
Stanford University Press, 1991.
---. Foundations of Cognitive Grammar: Theoretical Prerequisites.
Vol. I. Stanford, California: Stanford University
Press, 1987.
McGann, Jerome. Radiant Textuality: Literature after the
World Wide Web. New York: Palgrave, 2001.
Potter, R.G. “Statistical Analysis of Literature: A Retrospective.”
Computers and the Humanities 25.401-29
(1991).
Ramsay, Stephen. “Toward an Algorithmic Criticism.” Literary and Linguistic Computing 18.2 (2003): 167-74.
Rockwell, Geoffrey. “What Is Text Analysis, Really?”
Literary and Linguistic Computing 18.2 (2003): 209-19.
Rommel, Thomas. “Literary Studies.” A Companion to
Digital Humanities. Eds. Susan Schreibman, Ray Siemens
and John Unsworth. Oxford: Blackwell, 2004.
Rybicki, Jan. “Burrowing into Translation: Character Idiolects
in Henryk Sienkiewicz’s Trilogy.” Literary and
Linguistic Computing 26.1 (2005): 91-103.
Sinclair, Stefan. “Computer-Assisted Reading: Reconceiving
Text Analysis.” Literary and Linguistic Computing
18.2 (2003): 175-84.
Straznicky, Marta. Privacy, Play Reading, and Women’s
Closet Drama, 1550-1700. New York: Cambridge University
Press, 2004.
---. “Profane Stoical Paradoxes: The Tragedy of Mariam
and Sidney Closet Drama.” English Literary Renaissance
24 (1994): 104-34.
van Hoek, Karen. “Pronominal Anaphora.” Oxford
Handbook of Cognitive Linguistics. Eds. Dick Geeraerts
and Hubert Cuyckens. Oxford: Oxford University Press,
2007. 890-915.