Old English Universal Dependencies: Categories, Functions and

Specific Fields

Javier Martín Arista

Department of Modern Languages, University of La Rioja, E26004, Spain

Keywords: Universal Dependencies, Old English.

Abstract: The aim of this paper is to lay the foundations of the application of the framework of Universal Dependencies

to Old English. Such application will result in the morphological and syntactic annotation of a large data set

of Old English with Universal Dependencies categories and relations. The aim of this paper involves two

tasks. Firstly, it is necessary to select the relevant categories from the set of universal part-of-speech tags and

to identify the Old English exponents of the universal set of morphological features. Secondly, the dependency

relations holding in Old English should be listed. The main conclusion of this paper is that two specific fields

should be added to the standard Universal Dependencies annotation scheme in order to account to two central

aspects of Old English, namely, a gloss field, given the historical character of the language; and a

morphological relatedness field, in order to account for its associative lexicon.

1 INTRODUCTION

The aim of this paper is to lay the foundations of the

application of the framework of Universal

Dependencies (hereafter UD) to Old English. The

research reported here is the point of departure of the

morphological and syntactic annotation with UD of

ParCorOEv2 (henceforth ParCor), an open access

annotated parallel corpus Old English-English

(Martín Arista et al., 2021). Ultimately, the digitised

annotation of ParCor with structured data in a

relational database will allow for the computational

processing of Old English, including the specific

tools and techniques for low-resource languages

(Anastasopoulos 2019).

Universal Dependencies (de Marneffe et al.,

2021) is a model of morphological and syntactic

annotation devised for the compilation of

computerised data sets that facilitate cross-linguistic

comparison (de Marneffe et al., 2014) aimed to

natural language processing (Nivre, 2015) and to

areas of applied linguistics like language acquisition

and translation (MacDonald et al., 2013; Nivre,

2016). The annotation includes UPOS (universal part-

of-speech tags; Petrov et al., 2012), XPOS (language-

specific part-of-speech tags), Feats (universal

https://orcid.org/0000-0001-9900-0104

morphological features), lemmas, and dependency

heads and labels (Nivre et al., 2016). The 2015 release

of the UD dataset consisted of ten treebanks

representing ten languages, whereas the 2021 release

comprises 183 treebanks over 104 languages (Nivre

et al., 2020).

Old English is the historical stage of the English

language spoken in England between approximately

the 5

and the 11

centuries (CE). Written records,

which can be traced back to the 8

century onwards,

comprise approximately 3 million words in around

3,000 texts. The main lexicographical sources of Old

English include the dictionaries by Bosworth-Toller

(1973), Sweet (1976) and Clark-Hall (1996), as well

as the Dictionary of Old English (Healey, 2018). The

main textual sources of Old English are The

Dictionary of Old English web corpus (3,000,000

words; Healey et al., 2004) and The York-Toronto-

Helsinki Parsed Corpus of Old English Prose (Taylor

et al., 2003).

The paper is organised as follows. Section 2 and

Section 3 focus on the categorial part, while Section

4 deals with relational aspects. Section 5 presents the

extra fields required by Old English. Section 6 draws

the main conclusions.

Arista, J.

Old English Universal Dependencies: Categories, Functions and Speciﬁc Fields.

DOI: 10.5220/0010977300003116

In Proceedings of the 14th International Conference on Agents and Artiﬁcial Intelligence (ICAART 2022) - Volume 3, pages 945-951

ISBN: 978-989-758-547-0; ISSN: 2184-433X

945

2 PART-OF-SPEECH TAGS

UD annotation consists of categories and relations.

Beginning with categories, two types are considered:

lexical category and inflectional category. This is

called XPOS (language-specific Part-of-speech tags)

and Feats (universal morphological features). The

annotation also includes lemmas, dependency heads

and dependency labels (Nivre et al., 2016).

Table 1 compares the traditional set of parts of

speech and the universal part-of-speech set of UD

(Petrov et al., 2012). As can be seen in Table 1, the

set of UPOS is larger than the traditional inventory in

order to qualify as typologically valid. Notice that, in

Table 1, X stands for ‘foreign word’, sym is a non-

punctuation symbol and punct marks `punctuation’,

which is not used consistently in Old English but

rather added by text editors.

Table 1: Traditional parts of speech vs. universal part-of-

speech tags.

Traditional

POS

UPOS

Noun noun

propn

Verb verb

aux

Adjective adj

– det

num

Adverb adv

Pronoun pron

Preposition adp

Conjunction cconj

sconj

Interjection intj

– part*

– sym**

– punct***

Table 2 lists the specific categorial tags of Old

English. The right column in Table 2 provides

instances by category.

Table 2: Old English exponents and realisations of

Universal part-of-speech tags.

OE XPOS

realisations

common noun hlāfweard ‘steward’, mūða

‘mouth (of a river)’, sǣcol ‘jet’

proper noun Egipte ‘the Egyptians’,

Iringes weg ‘Milky Way’,

Legaceaster ‘Chester’

main verb sleacian ‘to slow’, ðoterian ‘to

cry’, twengan ‘to pinch’

auxiliary verb bēon ‘to be’, habban ‘to

have’, weorðan ‘to become’

adjective gnēað ‘frugal’, inwit ‘thick’,

meagol ‘mighty’

demonstrative-article se ‘the’

numeral (cardinal and

ordinal)

ðrīe ‘three’, fēowertīene

‘fourteen’, hundseofontigoða

‘seventieth’

adverb grundlinga ‘horribly’,

hedendlīce ‘strictly’,

pronoun sumhwilc ‘some’, ðu ‘you’,

ðīn ‘your’

adposition betweox ‘between’, gēan

‘against’,

coordinating

conjunction

and ‘and’, ge ‘and also’, oððe

‘or’

subordinating

conjunction

hwǣr ‘where’, ðēah

‘although’, ðȳ ‘because’

interjection ǣ ‘oh!’, ēuwā ‘wow!’, nū

‘lo!’

foreign word silua ‘forest’, torre ‘tower’

punctuation . , ; : ? !

3 MORPHOLOGICAL

FEATURES

In UD, morphological features have different values,

which are sorted alphabetically. The Old English

relevant features include pronominal type

ICAART 2022 - 14th International Conference on Agents and Artiﬁcial Intelligence

946

(demonstrative-article, indefinite pronoun,

interrogative pronoun, personal pronoun and relative

pronoun); numeral type (cardinal, ordinal);

possessive; foreign word; abbreviation; wrong

spelling; gender (feminine, masculine, neuter);

number (dual, plural, singular); case (accusative,

dative, genitive, instrumental, nominative); reflexive;

comparison (comparative, positive, superlative);

person (1, 2, 3); verbal form (finite, infinitive,

participe); mood (imperative, indicative,

subjunctive); tense (past, present); voice (active,

middle, passive); and polarity (affirmative, negative).

Nominal features like gender, number and case are

relevant to Old English verbs because present and

past participles are often inflected according to the

adjectival declension. On the features listed above, it

is also worth mentioning that wrong spellings are

supressed or normalised by editors, who frequently

provide alternative readings rather than ‘correct’ or

‘incorrect’ renderings. UD features and types as

applied to Old English are presented in Table 3.

Table 3: Features and values of Old English morphological

features.

Feature Values

ronominal t

e Dem-Art Ind Int Prs Rel

numeral t

e Card Or

ossessive Yes

foreign wor

Yes

abbreviation Yes

wrong spelling Yes

ende

Fem Masc Neut

numbe

Dual Plur Sin

case Acc Dat Gen In Nom

reflexive Yes

comparison Cmp Pos Sup

erson 1 2 3

verbal for

Fin Inf Part

moo

Ind Sub

tense Past Pres

voice Act Mid Pass

olarit

Aff, Neg

erson 1 2 3

ende

Fem Masc Neut

numbe

Dual Plur Sin

case Acc Dat Gen In Nom

invariable for

Yes

4 DEPENDENCY RELATIONS

The basic dependency relations distinguished in the

UD framework (de Marneffe et al., 2021) can be

broken down into nominal phrase dependencies,

simple clause dependencies and complex clause

dependencies. Dependency relations are defined by

means of dependency heads and labels. Dependency

heads are content words, while function words do not

usually show dependents of their own. The basic

opposition holds between core arguments (subjects,

objects and clausal complements) and oblique

modifiers (adjuncts and oblique arguments).

The following universal dependency relations are

found in Old English. acl (clausal modifier of noun),

acl:relcl (relative clause modifier), advcl (adverbial

clause modifier), advmod (adverbial modifier),

advmod:emph (emphasizing word, intensifier),

advmod:lmod (locative adverbial modifier), amod

(adjectival modifier); appos (appositional modifier),

case (case marking), cc (coordinating conjunction),

cc:preconj (preconjunct), ccomp (clausal

complement), csubj:pass (clausal passive subject),

conj (conjunct), cop (copula), csubj (clausal subject),

det (determiner), det:poss (possessive determiner),

discourse (discourse element), dislocated (dislocated

elements), fixed (fixed multiword expression), flat

(flat multiword expression), flat:foreign (foreign

words), flat:name (names), goeswith (goes with), iobj

(indirect object), list (list), mark (marker), nmod

(nominal modifier), nmod:poss (possessive nominal

modifier), nmod:tmod (temporal modifier), nsubj

(nominal subject), nummod (numeric modifier), obj

(object), obl (oblique nominal), obl:agent (agent

modifier), obl:arg (oblique argument), obl:lmod

(locative modifier), obl:tmod (temporal modifier),

orphan (orphan), parataxis (parataxis), punct

(punctuation), root (root) and vocative (vocative).

Although the dependency relations aux

(auxiliary), aux:pass (passive auxiliary), nsubj:pass

(passive nominal subject), expl (expletive) and

expl:impers (impersonal expletive) are relevant for

the syntactic annotation of Old English, it must be

borne in mind that they make reference to phenomena

that are on the grammaticalisation cline in Old

English (Denison 1993; Ringe and Taylor 2014; Petré

2014; Martín Arista and Ojanguren López 2018). For

instance, the Old English counterpart of the modal

auxiliary will can be used both as a general verb, in

instances like oððe hu he wolde ðæt hio wære… ‘or

how he would that it should be’

(BOET.005.046.013); and as a pre-auxiliary, as in

Forðam ic nu wille geornlice to Gode cleopian

‘Wherefore I will now earnestly call upon God’

(BOET.003.019.004). The passive has not been fully

grammaticalised in Old English yet, given that the

past participle frequently shows adjectival inflection

that expresses agreement with the subject, as is the

case in hi næron for nanum cræfte gecorene ‘they

were chosen for no virtue’, in which the masculine

Old English Universal Dependencies: Categories, Functions and Speciﬁc Fields

947

plural nominative gecorene agrees in gender, number

and case with the subject hie. There is still fluctuation

between bēon ‘to be’ and weorðan to become as

passive auxiliaries in Old English. As for the passive

subject, it is still possible for the agent to preserve the

dative case in the corresponding passive, thus And

him wæs gedemed fram unrihtwisum demum ‘And he

was judged by folly judges’ (ÆAdmon 1 006000

(4.31)). Regarding expletives, they are not

compulsory in Old English yet, as happens in Wæs

eac micel wundor þæt an wulf wearð asend ‘It was

also a great miracle that a wolf was sent’ (Æ LS

(Edmund) 003900 (145)), in which there is no

anticipatory subject to wæs. Finally, the formal

subject of impersonal verbs is often left unexpressed,

as in norþan sniwde ‘it rained from the north’ (Sea

000800 (31)). To summarise, the dependency

relations aux (auxiliary), aux:pass (passive auxiliary),

nsubj:pass (passive nominal subject), expl (expletive)

and expl:impers (impersonal expletive) are

distinguished in the annotation of Old English for

descriptive reasons, even though they have not been

fully grammaticalised yet.

5 OLD ENGLISH SPECIFIC

FIELDS

Once all the relevant categories and functions have

been identified or specified for Old English, the next

step is to decide whether additional tags (coded in

specific fields) are necessary to annotate this

historical stage of the language or not. The standard

CoNLL-U annotation format distinguishes ten fields,

listed and defined in Table 4.

Table 4: Fields in CoNLL-U annotation (from

https://universaldependencies.org/docs/format.htm).

ID Word index, integer starting at 1 for each

new sentence; may be a range for tokens

with multi

le words.

FORM Word form or

unctuation s

mbol.

LEMMA Lemma or stem of word form.

UPOSTAG Universal part-of-speech tag.

XPOSTAG Lan

e-s

ecific

art-of-s

eech ta

FEATS List of morphological features from the

universal feature inventory or from a

defined lan

e-s

ecific extension.

HEAD Head of the current token, which is either

a value of ID or zero (0).

DEPREL Universal Stanford dependency relation

to the HEAD (root iff HEAD = 0).

DEPS List of secondary dependencies (head-

rel

airs

)

MISC An

other annotation.

The tokenisation as well as the fields FORM and

LEMMA are imported automatically from ParCor.

Units smaller than a word (tokens) are presented in

the Appendix. They can also be imported from

ParCor. The fields UPOSTAG and XPOS are adapted

from the morphological tags of ParCor (see Appendix

on grammatical classes). At this stage, HEAD and

DEPREL are inserted manually.

Two extra specific fields are required to

adequately annotate the morphology and syntax of

Old English. The first is GLOSS. As we are

annotating a historical language, a translation into

Contemporary English facilitates the annotator´s task.

Inflectional forms, lemmas and glosses are

automatically imported from ParCor.

Table 5: Inflectional forms and glosses from ParCor.

Inflectional form Gloss

Ðu you

ðe who

ðam the

winterdagum winter days

selest givest

scorte short

tida times

and and

ðæs the

sumeres Summer´s

The second specific field proposed in this paper is

MORPHREL (morphological relatedness). In a

language characterised by the existence of large

derivational families with transparent morphological

relations (Kastovsky 1992) and generalised

inflectional inheritance of the prefix ge- (Martín

Arista 2012), a paradigmatic field specifying short-

distance and long-distance morphological relatedness

constitutes a remarkable explanatory resource. For

instance, the derived adjective unābrecendlic

‘inextricable’ is morphologically related (short-

distance) to the adjective *ābrecendlic as well as to

the primitive strong verb BRECAN ‘to break, tear,

crush, shatter, burst, break up, destroy, demolish’

(long-distance morphological relatedness). These

facts are indicated in the MORPHREL field as

*ābrecendlic / BRECAN.

6 CONCLUSION

This paper is the point of departure of the application

of the UD framework to the morphological and

ICAART 2022 - 14th International Conference on Agents and Artiﬁcial Intelligence

948

syntactic annotation of Old English. The relevant

categories and functions have been presented, while

some specific characteristics of the language have

been dealt with in terms of extra fields in the

annotation format. It remains for further research to

decide whether or not enhanced dependencies are

necessary to account for Old English null subjects and

objects, shared constituents in control and raising

constructions and antecedents of relative clauses.

ACKNOWLEDGEMENTS

Grant IPID2020-119200GB-100, funded by MCIN/

AEI/10.13039/501100011033/.

REFERENCES

Anastasopoulos, A. (2019). Computational Tools for

Endangered Language Documentation. Doctoral

dissertation, University of Notre Dame du Lac, Notre

Dame, Indiana, January.

Bosworth, J., Toller, T. N. (1973). An Anglo-Saxon

dictionary. Oxford: Oxford University Press.

Clark Hall, J. R. (1996). A concise Anglo-Saxon dictionary.

Toronto: University of Toronto Press.

de Marneffe, M., Manning, C., Nivre, J., & Zeman, D.

(2021). Universal Dependencies. Computational

Linguistics, 47 (2), 255–308.

de Marneffe, M., Dozat, T., Silveira, N., Haverinen, K.,

Ginter, F., Nivre, J., & Manning, C. (2014). Universal

Stanford Dependencies: a cross-linguistic typology.

Proceedings of LREC 2014 (pp. 4585–4592).

Denison, D. (1993). English Historical Syntax: Verbal

Constructions. London: Longman.

Healey, A. (Ed.) (2018). The Dictionary of Old English in

electronic form A-I. Toronto: Dictionary of Old English

Project, Centre for Medieval Studies, University of

Toronto.

Healey, A. (Ed.), Price Wilkin, J., & Xiang, X. (2004). The

Dictionary of Old English web corpus. Toronto:

Dictionary of Old English Project, Centre for Medieval

Studies, University of Toronto.]

Kastovsky, D. (1992). Semantics and vocabulary. In R.

Hogg (Ed.) The Cambridge history of the English

language I: The beginnings to 1066 (pp. 290–408).

Cambridge: Cambridge University Press.

Martín Arista, J. & Ojanguren López, A. E. (2018).

Grammaticalization and deflexion in progress. The past

participle in the Old English passive. Studia

Neophilologica 90(2): 155-175.

Martín Arista, J., Domínguez Barragán, S., García

Fernández, L., Ruíz Narbona, E., Torre Alonso, R., &

Vea Escarza, R. (comp.). 2021. ParCorOEv2. An open

access annotated parallel corpus Old English-English.

Nerthus Project, Universidad de La Rioja,

www.nerthusproject.com.

Martín Arista, J. (2012). The Old English prefix ge-: A

panchronic reappraisal. Australian Journal of

Linguistics, 32(4), 411–433.

McDonald, R., Nivre, J., Quirmbach-Brundage, Y.,

Goldberg, Y., Das, D., Ganchev, K., Hall, K., Petrov,

S., Zhang, H., Täckström, O., Bedini, C., Bertomeu, N.

& Lee, J. (2013). Universal Dependency Annotation for

Multilingual Parsing. Proceedings of the 51st Annual

Meeting of the Association for Computational

Linguistics (pp. 92–97).

Nivre, J., de Marneffe, M., Ginter, F., Hajiˇc, J., Manning,

C., Pyysalo, S., Schuster, S., Tyers, F., & Zeman, D.

(2020). Universal Dependencies v2: An evergrowing

multilingual treebank collection. Proceedings of the

Twelfth International Conference on Language

Resources and Evaluation (LREC 2020) (pp. 4027–

4036).

Nivre, J., de Marneffe, M., Ginter, F., Goldberg, Y., Hajiˇc,

J., Manning, C., McDonald, R., Petrov, S., Pyysalo, S.,

Silveira, N., Tsarfaty, R., & Zeman, D. (2016).

Universal Dependencies v1: a multilingual treebank

collection. Proceedings of the 10th International

Conference on Language Resources and Evaluation

(pp. 1659-1666).

Nivre, J. (2015). Towards a universal grammar for natural

language processing. In A. Gelbukh (Ed.),

Computational Linguistics and Intelligent Text

Processing (pp. 3–16). New York: Springer.

Nivre, J. (2016). Universal Dependencies: A Cross-

Linguistic Perspective on Grammar and Lexicon.

Proceedings of the Workshop on Grammar and

Lexicon: Interactions and Interfaces (pp. 38-40).

Petré, P. 2014. Constructions and environments: Copular,

passive, and related constructions in Old and Middle

English. Oxford: Oxford University Press.

Petrov, S., Das, D., & McDonald, R. (2012). A Universal

Part-of-Speech Tagset. Proceedings of the Eighth

International Conference on Language Resources and

Evaluation (LREC’12) (pp. 2089-2096).

Ringe, D., & Taylor, A. (2014). The development of Old

English. A linguistic history of English, vol. 2. Oxford:

Oxford University Press.

Sweet, H. (1976). The Student’s Dictionary of Anglo-

Saxon. Oxford: The Clarendon Press.

Taylor, A., Warner, A., Pintzuk, S., & Beths, F. (2003). The

York-Toronto-Helsinki Parsed Corpus of Old English

Prose [https://www-users.york.ac.uk/~lang22/Ycoe

Home1.htm].

APPENDIX

This appendix provides the full inventory of Old

English grammatical categories as well as a sample of

the abbreviations and compounds required for UD

token indexing.

Old English Universal Dependencies: Categories, Functions and Speciﬁc Fields

949

Demonstrative-Article

se ‘the’; þes ‘this’;

Relative pronouns

ðe ‘that, who, which

Interrogative pronouns

hūlic ‘of what sort’; hwā ‘who; what; any one, some

one, anything, something; each’; hwæðer ‘which of

two, whether.’; hwæðer...ðe each of two, both; one of

two, either’; hwæðer...or each of two, both; one of

two, either’; hwǣm ‘where’; hwæt ‘what’; hwæt

‘who; what; any one, some one, anything, something;

each’; hwilc ‘which, what; whosoever, whichever;

any (one), some (one)’; swā hwā ‘whosoever’; swā

hwā swā ‘whosoever’; swā hwæðer swā ‘whichever’;

swā hwǣr swā ‘wherever’; swā hwæt swā

‘whatsoever’; swā hwilc swā ‘whosoever’; swā...swā

‘whether...or; either...or’; swæðer ‘whichever,

whosoever’; tō hwǣm ‘wherefore

Indefinite pronouns

ǣghwā ‘each one, every one, everything, who or

whatever’; ǣghwæðer ‘everyone, either, both; each’;

ǣghwæt ‘each one, every one, everything, who or

whatever’; ǣghwǣt ‘anything’; ǣlc ‘each; any’; ǣnig

‘any, any one’; āhwa ‘any one’; āhwæðer ‘some one,

something; any one; anything; either, each, one or

other’; ānragehwā ‘everyone’; edwihte ‘anything,

something’; gehwā ‘each one, every one, any one,

whoever’; gehwæðer ‘both, either, each’; gehwilc

‘each, any, every (one), all, some, many, whoever,

whatever’; anra gehwilc ‘each one’; hwilcwega

‘some, any, someone; little, some, not much or great;

alone, anything, something’; ilca ‘the same’;

nāhthwæt ‘something unknown’; nāhwæðer

‘neither’; samhwylc ‘some’; sum ‘a certain one,

someone, something, one’; sumhwilc ‘a certain’; swā

‘so, the same, such, that’; swilc ‘such a one, he, the

same; such; (as a relative) which’; swilc...hwilc

‘such...as; so...as’; swilc...swilc ‘so much (many)...as;

as much (many)...as’; ðullic ‘such, such a’; welhwā

‘every one, every thing’; welhwilc ‘each, any, nearly

every

Personal pronouns

gē ‘you’; git ‘you two’; he ‘he, she, it; (pl.) they;

(reflex. pron.) himself, herself, itself’; hēo ‘she, they’;

hīe ‘they’; hit ‘it’; ic ‘I’; tū ‘thou’; ðæge ‘they, these’;

ðu ‘thou’; uncer ‘of us two, our (of two persons)’; we

‘we’; wit ‘we two

Coordinating conjunctions

ac ‘but; but also, moreover, nevertheless, however;

because, for’; ǣghwæðer ge...ge ‘both...and, as well

as’; ǣghwæðer...and ‘both...and, as well as’; and

‘and; but; or’; būtan ‘except, except that, but, only’;

būtan ðæt ‘except; unless, save that; except, but,

besides, if only, provided that’; eornostlīce ‘therefore,

but’; ge ‘and, also’; ge...ge ‘both... and; not only... but

also; whether... or’; nāhwǣðer ‘neither’; ne ‘neither,

nor’; oððe ‘or; and’; oððe...oððe ‘either...or’; ofðe

‘or’; sam ‘whether, or’; sam ðe...sam ðe

‘whether...or’; sam ge...sam ge ‘whether...or’;

sam...sam ‘whether...or’; samðe...samðe ‘as well...as

Subordinating conjunctions

ǣr ‘before that’; for ‘for, because’; for ðȳ ðe

‘because’; forðǣm ‘for (the reason) that, owing to

(the fact) that, for, because, on that account, therefore,

seeing that’; forðȳ ‘for that, because, therefore’;

forūton ‘without, besides, except’; gif ‘if; whether,

though’; hwǣðer ðe...ðe whether...or’; hwǣr ‘where,

whither, somewhere, anywhere, everywhere’; hweðer

‘whether’; hwȳ ‘why’; mid ðȳ ‘while, when’; nemne

‘unless, except, save, only’; nū ‘now that, inasmuch

as, because, since, when’; oð ‘until ‘; oððæt ‘until’;

sīð ‘after, afterwards’; siððan ‘as soon as, when,

since, after that, inasmuch as’; siððan...siððan

when...then’; sōðhwæðere ‘however, yet,

nevertheless’; swā ‘so as, consequently, just as, so far

as, in such wise, in this or that way, thus, so that,

provided that’; swā ðēah ‘nevertheless, yet.’; swilce

‘as if, as though’; tō ðȳ ðæt ‘for the purpose that, in

order that’; ðā ‘then, at that time; after that time,

thereupon; when, at the time that, whilst, during;

there, where; seeing that, inasmuch as, if, when,

since, as, because’; ðā ðā ‘when’; ðā hwīle ðe ‘while,

whilst, so long as’; ðā...ðā ‘then...when’; ðǣr ‘there,

thither, yonder; where, whither; then; when; though,

if, so far as, whilst, provided that; in that respect’; ðǣr

ðǣr ‘where, wherever’; ðǣr wið ‘in regard to that’;

ðǣr...of ‘therefrom’; ðǣrforan ‘before that’; ðæt ‘that,

so that, in order that, after that, then, thence’; ðæt ðe

‘that’; ðætte ‘that, so that, in order that. tō ðon ðætte

so that’; ðe ‘when; or; then; where. (with

comparatives) than’; ðe...ðe the...or, either...or’; ðēah

‘though, although, even if, that, however,

nevertheless, yet, still; whether’; ðēah ðe ‘although’;

ðēah... ðēah ‘although, still, yet’; ðenden ‘meanwhile,

while, as long as, until’; ðonne ‘then; therefore,

wherefore; yet; while, when; thereafter, henceforth;

rather than; since; although; (with comparatives)

than’; ðonne...ðonne ‘when...then’; ðonne...ðe ‘

‘since’; ðonne...gȳt ‘ ‘as yet, even’; ðonne...hwæðere

‘yet, nevertheless’; ðȳ ‘because, since, on that

account; therefore; then; (with comparatives) the’; ðȳ

lǣs ðe ‘lest’; ðȳ...ðȳ ‘the...the’; ðȳlǣs ‘lest’; weald ‘in

case’; weald ‘ðeah perhaps, possibly’; wið ‘until.

ICAART 2022 - 14th International Conference on Agents and Artiﬁcial Intelligence

950

Contractions by token (sample)

nage [ne+āgan], nagan [ne+āgan], nah [ne+āgan],

nanen [ne+ān], nanon [ne+ān], nanre [ne+ān], ]nænge

[ne+ǣnig], nænige [ne+ǣnig], nænigre [ne+ǣnig],

nære [ne+bēon], næaron [ne+bēon], næaron

[ne+bēon], nabban [ne+habban], næfden

[ne+habban], næfdon [ne+habban], nyllan

[ne+willan], nolde [ne+willan], nelt [ne+willan],

nitendum [ne+wītan], nyte [ne+wītan], nyten

[ne+wītan], naðer [nā+hwæðer], naðere

[nā+hwæðer], naðor [nā+hwæðer], nauht [nā+wiht],

nawht [nā+wiht], nawiht [nā+with], nanuht

[nān+wiht], nanwiht [nān+wiht], nanwit [nān+wiht],

nanwith [nān+wiht], nalæs [nā+lǣs], nallas [nā+lǣs],

nalles [nā+lǣs], nateshwan [nātes+hwōn], nateshwon

[nātes+hwōn], nahwider [nā+hwider], nahwanen

[nā+hwanon], næfre [ne+ǣfre], nahwider

[nā+hwider], nahwanen [nā+hwanon], næfre

[ne+ǣfre].

Compounds by token (sample)

ǣfenglōm / ǣfenglōma (noun) ǣfen + glōm

ilphlæden (adjective) gielp + hlæden

āncorlīf / āncerlīf (noun) āncor + līf

hringfāg / hringfāh (adjective) hring + fāg

hūhwega / hūhugu (adverb) hū + hwega

bedrēaf / beddrēaf (noun) bedd + rēaf

burgtūn / burhtūn (noun) burg + tūn

forelēoran / forlēoran (verb) fore + lēoran(ge)

hrǣwīc / hrēawīc (noun) hrǣw + wīc

ropwærc / hropwærc (noun) ropp + wærc

Old English Universal Dependencies: Categories, Functions and Speciﬁc Fields

951