CALLOE: A Pedagogical Tool for the Learning of Old English
1. Introduction: Existing Programs
For the last decades, CALL (Computer-Assisted Language Learning) software applications have been widely designed and marketed by both publishers (MacMillan, Routledge, CUP, OUP, etc.) and software companies (Citcom, DynEd, Edusoft, etc.). However, it is not strange that the number of such applications devoted to dead languages is rather limited. In the case of Old English, for instance, only a few specimens are available on the web, notwithstanding the heap of noteworthy digital electronic resources, especially electronic texts and editions, glossaries and bibliographic references (Foys 1999: 16-20; a good starting point might be http://www.supercable.es/~aalcaraz/HEL.htm, written in Spanish).
Among the most important contributions for Old English learning, we must highlight Cathy Ball's Hwæt! Old English in context (http://www.georgetown.edu/cball/hwæt_toc.html) which is basically a free on-line introductory course for beginners, organized around seven semantic fields, namely counting, nature, greetings, naming, months and seasons, fish, fowl and four-footed beasts, and the body along with a reference section and a glossary. Methodologically, each section has been designed to contain a set of related sentences and real texts for the user to read and translate, giving thus the chance to improve the knowledge of the Old English language as well as the Anglo-Saxon culture, "all within the constructs of a highly organized and easy to use website" (Bernstein 1997: 22). Morphology, however, happens to be beyond the scope of the course; a brief outline is a desideratum.
Jeremy J. Smith and Jean G. Anderson's The Essentials of Old English (http://www2.arts.gla.ac.uk/STELLA/ess.htm) is a licensed program designed to help students grasp the Anglo-Saxon language, as a part of the STELLA (Software for Teaching English and Scottish Language and Literature) project. The course has been planned in two parts, each consisting of six graded lessons with different types of exercises such as parsing, crosswords, comprehension exercises, gap-filling, text-adventure games, etc. (Conner et al. 1990: 32). As the authors point out in their webpage, the application has been thought of as a complement to the students' lessons, not for self-study. Texts are normalized into the West-Saxon standard as the course is designed for absolute beginners and dialectal variation may pose problems.
Alex Jones, from the University of Sydney, has implemented a tutorial program, English 2000, as a helpful tool for his students to translate some Old English texts (http://www.arts.usyd.edu.au/Arts/departs/english/2000ASNAC/OEtutes/OEtutes.html). The program is structured in ten units, along with a text, each one delving into a particular grammatical topic, but concentrating mainly (8 units) on morphology.
Old English Aerobics, a JavaScript based program designed by Peter Baker (http://www.engl.virginia.edu/OE/OEA), also deals with phonology, morphology and syntax. It includes a set of texts (The Wanderer, Wulf and Eadwacer, The Story of Caedmon, and the Life of Saint Æthelthryth) for the user to read and translate, offering helpful notes on grammar, meaning, idioms and clause type when you point out any word. It also features a complete range of exercises on all grammar topics.
Michael Drout's King Alfred is a software application especially designed to improve Old English through translation (Drout 1999: 29-34). The program has been organized from a two-fold perspective. First, users are provided with a set of sentences to translate into Modern English with a help assistant which may be clicked in case of doubt. This help mode offers thus information on all linguistic features such as part of speech, number, gender, tense, mood, and translation of the word needed, the latter being very helpful for pedagogical purposes as it allows the possibility of self-assessment.
Once a few sentences have been translated, the program makes use of a sorting algorithm to subsequently display on screen the three grammatical topics that the user has repeatedly failed, in order to review them. Such topics show a hyperlink giving access to the full explanation, paradigms, and morphological characterization of the word involved. Actually, one of the most attractive peculiarities of the program is that it allows the teacher to input the sentences to practice on, permitting then a focus on the structures and topics required.
2. Aims
The working system used in CALLOE (Computer-Assisted Language Learning of Old English) sets this program apart from those that we have briefly analyzed in the preceding section as it contains a morphological tagger, text library, sound, an exercise kit, and help modes along with the implementation of a True-Type font which allows the insertion of macrons and diacriticals. The present paper aims to describe CALLOE's applications, emphasizing those which have been proven to enhance the learning capability of our students.
The subject of Old English (OE) has been thoroughly affected by the changes adopted in the new educational system of Spain and other countries (Ritt 1999: 7) as the students of English philology are more and more frequently forced to learn OE with no previous knowledge of other highly inflected languages. This handicap, obviously, results in low motivation which becomes an added difficulty.
Although from our own experience we could give most of the main reasons which provoke the students' failure in OE, we considered it more trustworthy to check out our hypothesis against the students' opinion by handing out a questionnaire. For them, most problems arise from those features that distinguish OE from Present-Day English (high inflectional richness, grammatical gender, non-cognate vocabulary, different word-order, etc.). Due to its rich inflectivity, previous knowledge of Latin or German was considered an asset for the learning of OE declensions and conjugations as well as for the relationship between case and syntactic function. German vocabulary (and the 2nd consonant mutation) is also reported to be useful for OE lexis learning. Specific difficulties were also found when dealing with inflected infinitives and/or (absolute) participles, the distinction between strong and weak verbs/declensions, etc.
All these difficulties and others not listed here constitute some of the problems that we had to approach using computational linguistics. In order to ease the students' first steps in OE we have designed and implemented CALLOE, which is expected to assist them in the morphological analysis of Old English and in the understanding and translation of a text. For us it was a challenge to carry out this project which, at the same time, may convert a tiresome difficult learning subject into a motivating interactive activity.
3. Main features
CALLOE is a user-friendly application (to be run under a WINDOWS environment) which has been conceived with the pedagogical purpose of assisting the learning of OE. In this section we will analytically describe CALLOE by focusing on each of its components.
3.1. MAOET
CALLOE is provided with an automatic morphological analyzer and tagger of OE texts (MAOET) that we designed and implemented to satisfy our requirements after having previously studied other morphological taggers (Klavans and Chodorow 1991: 281-87; Antworth 1992: 389-98; Hogan-Brun and Whittle 1998: 451-57; see also Wolff 1998; Cazade 1999). MAOET performs the morphological analysis of each word by dividing the items into the three possible chains of characters in which a word may be split for its morphological analysis and, after consulting the corresponding dictionaries (prefixes, lexemes, and morphemes) and applying a context-free grammar, it subsequently displays the corresponding tagging on screen.
As mentioned above, MAOET works with a context-free grammar defined by a set of rules that are used to connect the lexemes with their corresponding morphemes. These rules state which "lexeme continuation class" can be followed by which "morpheme continuation class," a task performed by matching the contents of a common key field in both dictionaries. The threefold dictionary structure and the grammar employed has turned out to be very efficient on account of the tests carried out as MAOET reliably analyzes and tags a great number of words without increasing the size of the lexical dictionary at an average rate of fifteen words per second using a microprocessor at 200 MHz.
Once a text has been loaded, CALLOE offers the possibility of analyzing the text sentence by sentence or as a whole. Once the text has been processed, it displays the tagging which can be scrolled for consultation, saved as a text file, or printed. If the first option is chosen, the following sentence may be analyzed at the user's request and the same procedure may be repeated until the end of the text. Instead of a sequential analysis, the user may also follow a random procedure by just highlighting the words that he/she wants to be analyzed.
As an illustration, we reproduce the tagging of the following text:
Hī swīðe irlīcum andwlitan beseah tō ðæm iungan ealdormen and cwæð: 'Þū iunga mann, canst ðū þone dōm mīnre dohtor gifta?' (The Old English Apollonius of Tyre, Goolden 1958: iv, 6-8).
Hī
[hī:PronPersSing3rdNomMasc<>he} ]
swīðe
[swīð:AdjeStrongPlurAccFeme<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongPlurAccMasc<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongPlurAccNeut<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongPlurNomFeme<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongPlurNomMasc<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongPlurNomNeut<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongSingAccFeme<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongSingInstMasc<>strong,powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongSingInstNeut<>strong,powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongSingNomFeme<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongSingNomMasc<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeStrongSingNomNeut<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeWeakSingAccNeut<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeWeakSingNomFeme<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīð:AdjeWeakSingNomNeut<>strong, powerful; (comp. swīðra) right (hand)} ]
[swīðan:VerbWeak#1Sing1stPrsInd<>strengthen; (ofer-) overcome, conquer} ]
[swīðan:VerbWeak#1Sing2ndImper<>strengthen; (ofer-) overcome, conquer} ]
[swīðan:VerbWeak#1SingAllPrsSub<>strengthen; (ofer-) overcome, conquer} ]
[swīðan:VerbWeak#2Sing1stPrsInd<>strengthen; (ofer-) overcome, conquer} ]
[swīðe:Adve<>very} ]
irlīcum
[irlīc:AdjeStrongPlurDatFeme<>angry} ]
[irlīc:AdjeStrongPlurDatMasc<>angry} ]
[irlīc:AdjeStrongPlurDatNeut<>angry} ]
[irlīc:AdjeStrongSingDatMasc<>angry} ]
[irlīc:AdjeStrongSingDatNeut<>angry} ]
[irlīc:AdjeWeakPlurDatFeme<>angry} ]
[irlīc:AdjeWeakPlurDatMasc<>angry} ]
[irlīc:AdjeWeakPlurDatNeut<>angry} ]
andwlitan
[andwlita:NounWeakPlurAccMasc<>face} ]
[andwlita:NounWeakPlurNomMasc<>face} ]
[andwlita:NounWeakSingAccMasc<>face} ]
[andwlita:NounWeakSingDatMasc<>face} ]
[andwlita:NounWeakSingGenMasc<>face} ]
beseah
[sēon:VerbStrong#5SingFirst & 3rdPrtInd<be>see,look: (for-) despise, scorn} ]
tō
[tō:Adve<>too} ]
[tō:PrepReg/Gen/Acc/Dat<>to & towards & as & for, with a view to} ]
ðæm
[se, sēo, þæt:PronDemoPlurDatEpic<>the; that, which, what} ]
[se, sēo, þæt:PronDemoSingDatMasc<>the; that, which, what} ]
[se, sēo, þæt:PronDemoSingDatNeut<>the; that, which, what} ]
iungan
[geong:AdjeWeakPlurAccFeme<>young} ]
[geong:AdjeWeakPlurAccMasc<>young} ]
[geong:AdjeWeakPlurAccNeut<>young} ]
[geong:AdjeWeakPlurNomFeme<>young} ]
[geong:AdjeWeakPlurNomMasc<>young} ]
[geong:AdjeWeakPlurNomNeut<>young} ]
[geong:AdjeWeakSingAccFeme<>young} ]
[geong:AdjeWeakSingAccMasc<>young} ]
[geong:AdjeWeakSingDatFeme<>young} ]
[geong:AdjeWeakSingDatMasc<>young} ]
[geong:AdjeWeakSingDatNeut<>young} ]
[geong:AdjeWeakSingGenFeme<>young} ]
[geong:AdjeWeakSingGenMasc<>young} ]
[geong:AdjeWeakSingGenNeut<>young} ]
ealdormen
[man(n):NounPlurAccMasc<ealdor>man} ]
[man(n):NounPlurNomMasc<ealdor>man} ]
[man(n):NounSingDatMasc<ealdor>man} ]
and
[and & ond:ConjCopu<>and} ]
cwæð
[cweðan: VerbStrong#5SingFirst & 3rdPrtInd<> say; (be-) bequeath; (wiþ-) speak against, oppose, deny} ]
þū
[þū:PronPersSing2ndNom<>thou (you)} ]
iunga
[geong:AdjeStrongPlurAccFeme<>young} ]
[geong:AdjeStrongPlurAccMasc<>young} ]
[geong:AdjeStrongPlurNomFeme<>young} ]
[geong:AdjeWeakSingNomMasc<>young} ]
mann
[man(n):NounSingAccMasc<>man} ]
[man(n):NounSingNomMasc<>man} ]
[mannian:SimbWeakSing2ndImper<>man, garrison} ]
canst
[cunnan:VerbAnomSing2ndPrsInd<>know} ]
ðū
[ðū:PronPersSing2ndNom<>thou (you)} ]
þone
[ se, sēo, þæt:PronDemoSingAccMasc<>the; that, which, what} ]
[þonne:AdveTime<>then} ]
dōm
[dōm:NounStrongSingAccMasc<>judgement, decree} ]
[dōm:NounStrongSingNomMasc<>judgement, decree} ]
[dōn:VerbAnomSing1stPrsInd<>do; (ge-) make, act, behave; (for-) destroy} ]
mīnre
[mīn:AdjeStrongSingDatFeme<>my} ]
[mīn:AdjeStrongSingGenFeme<>my} ]
[mīn:PronPossSing1stDatFeme<>my} ]
[mīn:PronPossSing1stGenFeme<>my} ]
dohtor
[dohtor:NounPlurAccFeme<>daughter} ]
[dohtor:NounPlurNomFeme<>daughter} ]
[dohtor:NounSingAccFeme<>daughter} ]
[dohtor:NounSingDatFeme<>daughter} ]
[dohtor:NounSingGenFeme<>daughter} ]
[dohtor:NounSingNomFeme<>daughter} ]
gifta
[gift:NounStrongPlurAccFeme<>marriage gift,dowry; (pl.) nuptials, marriage} ]
[gift:NounStrongPlurGenFeme<>marriage gift,dowry; (pl.) nuptials, marriage} ]
[gift:NounStrongPlurGenNeut<>marriage gift,dowry; (pl.) nuptials, marriage} ]
[gift:NounStrongPlurNomFeme<>marriage gift,dowry; (pl.) nuptials, marriage} ]
As shown, the tagging contains all the possible right analyses for each word, leaving to the students the task of disambiguation which turns out to be very instructive as they must choose the correct analysis in view of the context.
3.2. Text library
CALLOE contains a library of OE texts to be analyzed by MAOET. Most of them are pedagogically adapted OE texts for beginners or intermediate level learners (Brook 1955; de la Cruz 1986; Mitchell 1995; de la Cruz et al. 1995). Along with them, The Old English Apollonius of Tyre is also available for analysis. The texts are marked with the macron to indicate vowel length as we think that the use of the diacritic becomes essential to know the meaning of many OE words (for instance, god vs. gōd, ge vs. gē, witan vs. wītan, etc.).
The library is open to include other texts which may be written by the user in the EDIT facility or imported from other sources; and a specially-created True-Type font (FONETIK) has been designed for special characters (macron and IPA). This facility allows the student to practise individually with single words chosen by him/herself.
3.3. Sound device
CALLOE is supplied with sound files which reproduce some texts, the readings of which have been previously stored. This feature offers the students the possibility of alternatively reading and listening to the text, thus realizing the main characteristics of the OE sound system.
3.4. Help modes
CALLOE is supplied with a Help mode at the users' disposal to give general assistance information about the program and about Old English, activated by just clicking on the Help window. The next unfolded windows offer all the possibilities from which to choose. The specific information can be summarized as follows:
a) Morphology. Within this option, paradigmatical information about nouns, verbs, adjectives, pronouns, etc., is provided and, when necessary, characters written in different colours are used to distinguish between lexeme (stem) and morpheme (inflection). For each of the paradigms further windows are unfolded with a tree-branch format. For example, in the case of verbs the students are led into strong, weak, preterite-present and anomalous verbs. In the first option tense formation, classification, strong verbs classes, etc. is offered. Finally, for each of the seven classes an alphabetical list of the main verbs is displayed, including infinitive, 3rd person singular present indicative, preterite1, preterite2, past participle, and meaning. Similar treatment is offered for the other paradigms.
b) Phonology and orthography. To illustrate the relationship between OE phonemes (or allomorphs) and graphs, we use several tables containing monophthongs and diphthongs as well as consonantal sounds and their corresponding (di)graphs. Examples with their phonemic notation (IPA) are also provided.
c) Translation. To help the students solve their doubts and check their own translations, a Spanish version of the adapted texts is available.
3.5. Exercise-Manager
CALLOE is supplied with an exercise manager which allows the students to drill in OE morphology (any part of speech or as a whole) just by supplying the right form from its morphological analysis.
4. Final remarks
Our program is in constant updating and new, more refined prototypes are launched on a regular basis. Most important, the number of lemmas in the root dictionary is being increased as new texts (both in prose and verse) are being tagged. In our next version, we will also offer some excerpts from the OE Gospels along with some Exeter Riddles and a tagged version of The Wanderer. Furthermore, new sound files will be recorded for some other texts, especially those written in verse.
We expect to supplement the Help modes by including further paradigms as well as giving a short overview on some particular topics on OE syntax, phonology, and metrics; an English version of the User's Guide will also be available shortly. Additionally, the Exercise mode will also be improved so as to allow variation and gradation (more varied exercises and drills are required). Half-cloze exercises (providing the lemma but not the inflexional morphemes) will be offered for the users to fill in accordingly.
CALLOE can be downloaded from the following Webpage http://alcor.lcc.uma.es/~trivino/calloe/. It is freeware, although you have to register in advance, as the program will ask you for an access number. Simply click on Formulario de registro y descarga de CALLOE and follow the steps. Once the information is sent, you will receive an e-mail with your personal identification number. We will sincerely appreciate all the suggestions that the user may make. We will attempt to take them into consideration in future versions of our program.
5. References
- Antworth, E.L. "Glossing Text with the PC-KIMMO Morphological Parser." Computers and the Humanities 26.5-6 (1992): 389-398.
- Bernstein, M.J. "An Introduction to the Internet for Anglo-Saxonists. Part 2: Pedagogical and Scholarly Highlights." Old English Newsletter 31.1 (Fall 1997): 22-27.
- ----. "New and Updated Anglo-Saxon Resources on the Internet for 1998." Old English Newsletter 32.1 (Fall 1998): 14-15.
- Brook, G.L. An Introduction to Old English. Manchester: Manchester University Press, 1955.
- Calle Martín, J. and J.L. Triviño Rodríguez. "Algoritmos de Derivación de Palabras con Ortografía Irregular en el Análisis Morfológico Automático del Inglés Antiguo." Interlingüística 9 (1998): 67-70.
- Cazade, A. "'Time and Time Again': A Response to Dieter Wolff on CALL." The European English Messenger 8.1 (1999): 67-70.
- Conner, P.W., M. Deegan and C.A. Lees. "Computer-Assisted Approaches to Teaching Old English." Old English Newsletter 23.2 (Winter 1990): 30-35.
- de la Cruz, J. Iniciación Práctica al Inglés Antiguo. Madrid: Alhambra, 1986.
- de la Cruz, J., Cañete, A., and Miranda, A. Introducción Histórica a la Lengua Inglesa. Málaga: Ágora, 1995.
- de la Cruz, J., A. Cañete and A. Miranda. Textos y Vocabulario para la Introducción Histórica a la Lengua Inglesa. Málaga: Ágora, 1995.
- Foys, M. "Circolwyrde 1999: New Electronic Resources for Anglo-Saxon Studies." Old English Newsletter 33.1 (1999): 16-20.
- Goolden, P. The Old English Apollonius of Tyre. Oxford: Oxford University Press, 1958.
- Hogan-Brun, G. and R. Whittle. "The Potential of Multi-Media for Foreign Language Learning: A Critical Evaluation." Computers and the Humanities 31 (1998): 451-457.
- Klanvans, J.L. and M.S. Chodorow. "Using a Morphological Analyzer to Teach Theoretical Morphology." Computers and the Humanities 25 (1991): 281-87.
- Mitchell, B. An Invitation to Old English and Anglo-Saxon England. Oxford: Blackwell, 1995.
- Ritt, N. "Historical Linguistics in English Studies: Problems and Perspectives." The European English Messenger 8.1 (Spring 1999): 7-8.
- Sproar, R. Morphology and Computation. Bradford: ACL-MIT Press, 1992.
- Wolff, D. "Computers and New Technologies in Language Learning." The European English Messenger 7.2 (1998): 69-71.
