Babylon: Displacement and Recreation of Calderón's Life is a Dream

With the emergence of digital texts and the movement
to digitize existing texts, a complementary
push to develop digital means of textual analysis continues
to surface. New digital analysis tools can draw
on large databases of texts or apply multiple analysis
methods to a single text in minutes. Thus, digital textual
analysis potentially provides the opportunity to examine
large amounts of texts in scant time (Moretti, 2007),
while offering visualization potentials that exceed that of
texts. For example, these tools can create graphs of word
occurrences, map those occurrences according to concordance,
or remove all occurrences of a word to show
its frequency in a text.
Described by Rockwell (2003) as exploring “the question
of the relationship between how we represent texts,
how we see them, and our theories of textuality,” textual
analysis generally seeks to identify patterns within
the text, such as concordance or unity (Rockwell, 2003),
meaning (Samuels & McGann, 1999), truth (Brooks,
1947), or rhetorical strategy (Bazerman & Prior, 2004).
Yet, another implication of digital textual analysis is that
it involves the re-creation of a text. This provides an opportunity
for us to develop a heightened understanding
of what is the role of readers and writers as co-creators of
literature, and establishes the theoretical basis necessary
to link textual analysis with translation. As its Latin root
translatio (to transfer, to carry, to bring across, to displace)
suggests, translation’s basic function is to move
meanings from one context to another; consequently,
translation can denote concepts such as paraphrase,
decoding, interpretation, communication, and even recreation.
Borges’ theory of translation, as illustrated in his famous
works “Pierre Menard, Author of the Quixote” and
“The Translators of the One Thousand and One Nights,”
opens up a world of limitless possibilities by elevating
translation from an act of reproduction to one of re-creation.
In the fictional short story Pierre Menard, Borges
highlights the importance of the context, the reader and
the translator as co-creators of meaning, ushering a new
theory of literary criticism and translation that trivializes
the preoccupation with notions such as faithfulness, authorship
and originality. He supports this theory in his
essay on the translators of the Arabian Nights, by making
the translator’s infidelities into what is most important
and valued. As a result, he demonstrates that literary
translations can produce diverse representations of the
foreign text and culture, which ultimately enrich both
languages and texts through the act of remaking.
Utilizing Borges’ theory as the conceptual backbone, this
project blends traditional and digital methods of textual
analysis to create visualizations (tag clouds, word trees,
influential word maps) that serve as entry points for
analysis. With the goal in mind of investigating translation
theory, the paper studies the effects of man-made,
machine-made translations, and digital text-analyses of
a passage from Spanish Golden Age drama. This approach
provides a greater understanding of how Spanish
source editions (Cruickshank, and Williamsen), manmade
translations (Racz, FitzGerald, and MacCarthy),
and machine-made translations (Babylon software) versions
of a soliloquy from Calderón’s play “La vida es
sueño,” compare and contrast to one another. To re-create
both source and man-made translation texts the Babylon
translation software tool was applied, first to convert
them from Spanish into English, or English into Spanish,
and then to translate those versions back into the original
language, Spanish or English. After completing the
first level operation of running the source text repeatedly
and recursively through the Babylon translation software
application, which utilizes a statistical approach to automatic
language translation and natural language processing,
the second level operation was set in motion.
This level involved decomposing all online texts (source
editions, man-made translations, machine-made translations)
into their constituent elements and patterns with
the aid of text-analysis and statistical visualization tools.
In order to compare and contrast all versions amongst
themselves and with each other, various text-analysis
tools, namely TagCrowd, TAPoR, ManyEyes and Crawdad
were utilized to visualize word patterns, themes,
and word frequencies along with their importance. It is
significant to note that the tools trended toward certain
forms of interpretation, and that these trends affected the
types of analysis possible, but, more importantly, they
could also be re-purposed for alternative approaches.
The two tools that were found to be the most useful for
the comparison of different versions, as well as for their
application to Spanish texts, were ManyEyes and Crawdad.
ManyEyes was employed to generate tag clouds
and word trees. While tag clouds were used to visualize
word frequencies with number of occurrences in their
context, which is useful in comparing word usage patterns; word trees were particularly helpful in visualizing
individual words, phrases and punctuation concordances
show within their respective context in order to
reveal recurrent themes. Crawdad, on the other hand,
was employed to visualize a network of the most influential
words and their interconnections, along with their
frequency of co-occurrence. The tool also provided influence
scores, which served as indicators of high coherence
and focus in the text.
The overall intent of the project is not to determine which
version is more faithful to the “original,” but rather to
explore how these creative “infidelities” and re-creations
enrich both languages and cultures. Through translation
and digital textual analysis we can illuminate the challenges
and possibilities presented by the transformation/
displacement of seemingly discrete national, cultural and
literary territorialities. For example, of particular interest
in the research results is the appearance of deviations
in minor themes, which demonstrate that, although, main
themes predictably correlate closely across versions,
subordinate themes follow less predictable patterns. By
studying these variations we can explore how each recreation
reflects the intersection between languages, as
well as the existence of latent possibilities of diverse
meaning contained within the source text. The result
is the development of a networked fidelity framework,
which rejects the traditional faithful/unfaithful binary
that has predominated in translation studies.
The project corroborates the impossibility of a word for
word translation and even edition, and further supports
the acceptance of multiple versions as all simultaneously
genuine and divergent, while highlighting the difficulty
and futility of placing versions in a hierarchy. Additionally,
since all translations must be equally encompassed
as “unfaithful,” they can also be judged as “faithful,”
highlighting the fact that translation is an act of subjective
interpretation and re-creation. Consequently, translation
(be it machine or man-made) shares similarities
not only with text-analysis, but also with deformance,
and the profound subjectivity from which critical insight
emerges. This posits that translations not only have the
power to share the work with a wider culture, but also to
enrich both languages and texts through the act of remaking.
This, in turn, establishes the possibility of equating
translation to a creative process that results in unique
text, eliminating the hierarchical divide that exists between
author and translator, and original and translation,
as envisioned by Borges’ theory of translation.
Furthermore, the project shows that digital visualization
tools for “distant” reading and traditional methods of
close reading do inform and complement each other to
expand our understanding of all texts, including those
in translation. However, do to their dependence on each
other they do require flexibility in order to switch back
and forth between both modes of interpretation. Lastly,
there are many possible ramifications of this research
project. For example, the use of machine-assisted translation
for the analysis (and interpretation) of translations,
not just for the actual generation of translated
texts, as well as an alternative way to explore how meaning
is constructed and re-constructed in literature.
Works Cited
Babylon: Translation at a single click. 1997-2007. Babylon
Ltd. 13 Feb. 2008. <http://www.babylon.com/>.
Bazerman, Charles and P. Prior What Writing Does and
How it Does it. Mahwah, New Jersey: Lawrence
Earlbaum Associates, 2004.
Borges, Jorge Luis. “Pierre Menard, Author of the Quixote.”
Ficciones. English Trans. Buenos Aires: Editorial
Sur, 1944. 88-95.
- - - . “The Translators of the One Thousand and One
Nights.” Trans. Esther Allen. The Translation Studies
Reader. Ed. Lawrence Venuti. 2nd ed. New York: Routledge,
2004. 94-108.
Brooks, Cleanth The Well Wrought Urn: Studies in the
Structure of Poetry. Orlando, Florida: Harcourt Brace &
Company, 1947.
McCarty, Willard, Humanities Computing New York:
Palgrave MacMillan, 2005.
Moretti, Franco. Graphs Maps Trees: Abstract Models
for a Literary History. London: Verso, 2007.
Samuels, Lisa Jerome McGann. “Deformance and Interpretation.”
New Literary History, 30.1 (1999): 25-56.
Rockwell, G. “What is Text Analysis, Really?”