Association for Computers and the Humanities

Poster Sessions from ALLC/ACH '96

Palimpsest for Windows

A Tool for Computer Assisted Literary Translation Studies

Encoding	Viewing and Editing
Several editions of the original text and its translations in ASCII- or SGML-endcoded form	Preprocessed corpus
Wordprep generates word-based tables Output: dBASE IV tables	The Viewer provides a multilingual,multilateral phrase-aligned interlinear display of the texts, including searching facilities for a word and a lemma. The Editor allows for easy correlation of translated and original elements.
Phrprep generates phrase-based tables Output: dBASE IV tables
Concordancing	Statistics
Preprocessed corpus	Preprocessed corpus
Automatic generation of word- or phrase-based and -aligned concordances; word-based concordances include the phrase in which the word occurs. Output: dBASE IV tables	Automatic generation of word- or phrase-order statistics with optional sentence alignment (ie. word- or phrase order relative to the corresponding sentences rather than to the entire text) Output: dBASE IV tables
Access 2.0 report: Conversion of Palimpsest results into a Word for Windows compatible text file	Excel 5.0 diagrams: Conversion of Palimpsest data into two or three dimensional graphs

Alignment:

Correlated elements:	word, phrase
Criteria for correlation:	* lexical congruence: dictionary "equivalents"
	* "logical" congruence: textual "equivalents"
Full, explicit correlation:	translation -> original
Partial, implicit correlation:	original -> translation, translation -> translation

Excerpt from a word-based table:

ID	Reference	Wordtag	Word	Lemma	Parse	Link	Phrase
8	T001a:0000035	ART	Das	das	[S [.	35	7
9	T001a:0000040	NN	Meer	Meer	^]	40	7
10	T001a:0000045	VFIN	war	sein	[^]	45	8
11	T001a:0000050	APPRART	vo	von	[^	55	9
12	T001a:0000051	APPRART	+m	der	[.	60	9
13	T001a:0000055	NN	Himmel	Himmel	^] ]	65	9
14	T001a:0000060	PTKNEG	nicht	nicht	[.	50	11
15	T001a:0000065	PTKZU	zu	zu	.	50	11
16	T001a:0000070	VINF	unterscheiden	unterscheiden	^]	50	11
17	T001a:0000075	$,	+,		.		6
18	T001a:0000080	PPER	es	es	[S- [^]	85	13
19	T001a:0000085	VFIN	war	sein	[^]	95	14

WAWORD1.DBF (The Waves translated by Herberth and Marlys Herlitschka, 1959)

Excerpt from a phrase-based table:

ID	Startp	No	Tag	Text	Lemtext	L. 1
7	8	7		Das Meer	das Meer	7
8	10	8		war	sein	8
9	11	9		vom Himmel	von Himmel	10
10	14	11		nicht zu unterscheiden	nicht zu unterscheiden	9
11	18	12	#S-			12
12	18	13		es	es	14
13	19	14		war	sein	15
14	20	15		nur	nur	13
15	21	16		ein wenig gefältelt	ein wenig gefältelt	16

WAPHR1.DBF (The Waves translated by Herberth and Marlys Herlitschka, 1959)

Excerpts from the phrase-based concordance:

a blue finger-print. of. shadow	a broad flame
einen blauen Fingerabdruck. von. Schatten	eine breite Flamme
einen blauen Schattenfingerabdruck	eine breite Flamme
einen blauen Fingerabdruck. aus Schatten	eine breite Flamme
d'ombres bleues	une large flamme
une empreinte. d'ombre bleue	une large flamme
a million atoms. of. soft blue	on. the rim. of. the horizon
in. Millionen Atome. von. sanftestem Blau	am Himmelsrand
in. eine Million. hellblauer Atome	am Rande. des Horizontes
zu. einer Million. sanft blauer Atome	am Rand. des Horizonts
en. un million. d'atomes bleu tendre	sur. le rebord. du ciel
en. millions. d'atomes bleu pâle	au bord. de. l'horizon

The data relates to

The Waves by Virginia Woolf (1931),
Die Wellen trans. Marlys and Herberth Herlitschka (1959),
Die Wellen trans. Marie Bosse-Sporleder (1994),
Die Wellen trans. Jan-Mirko Maczewski (1995),
Les Vagues trans. Marguerite Yourcenar (1937), and
Les Vagues trans. Cécile Wajsbrot (1994) respectively.

Palimpsest generated graph displaying intra-translational phrase order relations

Y1 = H. and M. Herlitschka; Y2 = M. Bosse-Sporleder; Y3 = J-M. Maczewski
Y4 = M. Yourcenar, Y5 = C. Wajsbrot
X: phrase number; Y: ABS(phrase number (translation A) - phrase number (translation B))

Technical Information

The experimental corpus:

The word- and phrase-aligned corpus is a "collective text" consisting of the first italicized passage of Virginia Woolf's novel The Waves and its two French and three German translations. The word-based tables consist of approximately 450 reco rds, the phrase-based tables of approximately 210 records each. As a whole, the preprocessed corpus occupies nearly 2 MB of hard disk space. The sources of the texts are given below.

The original text
Woolf, Virginia (1992). The Waves. London: Penguin.

The French Translation
Les Vagues (1937), trans. Marguerite Yourcenar. Paris: Éditions Stock.
Les Vagues (1993), trans. Cécile Wajsbrot. Paris: Calmann-Lévy.

The German Translations
Die Wellen. Roman. (1979), trans. Herberth and Marlys Herlitschka. Frankfurt: Fischer Taschenbuch Verlag.
Die Wellen. Roman. (1994), trans. Maria Bosse-Sporleder. Frankfurt: Fischer Taschenbuch Verlag.
"Die Sonne war noch nicht aufgegangen." (1995), trans. Jan-Mirko Maczewski. Appendix 6.1 to the MA thesis. Colchester: University of Essex.

System requirements:

CPU: 486 DX 33 or faster
RAM: 4 MB or more
Operating System: Windows 3.1 or higher
Free hard disk space: 10 MB
Additional software for preprocessing: Any editor capable of handling dBASE-files
Recommended software for postprocessing: Access 2.0 or higher, Excel 5.0 or higher

ACH Logo Head Go back to the ACH Home Page

Any questions or bug reports regarding the ACH Web Pages should go to ACHWeb@brown.edu