Elamur: **Mova** · [Parse](https://mova.elamur.ai/free/en/parse/) · [Dictionary](https://md.dictionary.elamur.ai/free/en/) · [Translate](https://md.translate.elamur.ai/free/en/) · [Research](https://md.research.elamur.ai/free/en/) · [Gali](https://md.gali.elamur.ai/free/en/) · [MCP](https://mcp.elamur.ai/) · [Locations](https://md.locations.elamur.ai/free/en/) · [Weather](https://md.weather.elamur.ai/free/en/) · [Constitution](https://md.constitution.elamur.ai/)

[Mova](https://md.mova.elamur.ai/free/en/) · [this page in HTML](https://mova.elamur.ai/free/en/docs/languages/is/non/01-introduction/)

[Mova](https://md.mova.elamur.ai/free/en/) / [docs](https://md.mova.elamur.ai/free/en/docs/) / [languages](https://md.mova.elamur.ai/free/en/docs/languages/) / [is](https://md.mova.elamur.ai/free/en/docs/languages/is/) / [non](https://md.mova.elamur.ai/free/en/docs/languages/is/non/) / 1. Old Norse and the conventions of this manual

# 1. Old Norse and the conventions of this manual

Old Norse is the medieval North Germanic language known from runic inscriptions and manuscripts. This manual serves the `non` edition and uses normalized Old West Norse for its main paradigms. It identifies the variety and the form of the evidence whenever either affects the analysis.

## North Germanic affiliation and communities of use

Old Norse belongs to the North Germanic branch of Germanic, within Indo-European. Its earlier North Germanic antecedent is conventionally called Proto-Norse. The change from Proto-Norse to Old Norse was gradual, and the dating of the boundary depends on which sound changes, inflections or inscriptions are used to identify it. The Viking Age and medieval evidence also spans regional differences that grew more visible over time. (Noreen 1923, § Einleitung.)

North Germanic speakers lived across Scandinavia and in settlements overseas. The western varieties are associated chiefly with Norway, Iceland and the North Atlantic settlements, including the Faroes and Greenland. Norse was also spoken in parts of the British Isles. The eastern varieties are associated chiefly with Denmark and Sweden. These names describe broad geographical groupings. A text’s place of composition, a copyist’s variety and the place where a manuscript was preserved can each supply different evidence about its language. (Noreen 1923, § Einleitung.)

The surviving record includes short inscriptions, laws, narrative prose, verse, religious writing and learned texts. Runic and manuscript evidence overlap in date. The abundance of Icelandic manuscripts gives Old Icelandic particular weight in the description of medieval grammar. The distribution of surviving texts is evidence about what was written, copied and preserved. A claim about speech in a particular community requires evidence that can be connected to that community. (Noreen 1923, § Einleitung; Sweet 1886, § Preface.)

Old Norse is a historical language. Its varieties continued into later North Germanic languages, including Icelandic, Faroese, Norwegian, Danish and Swedish. Modern readers can learn the medieval language through normalized editions, and some modern languages preserve many of its forms. The grammatical descriptions here concern medieval usage. (Noreen 1923, § Einleitung.)

## The scope of the non edition

`non` is the single edition code served by this manual. The code identifies the language at the edition level. Within the edition, names such as Old Icelandic and Old Norwegian identify varieties, and the date and source of an example identify the evidence for a particular use. Forms from Old Danish and Old Swedish enter the description where eastern comparison clarifies a distribution or where a passage requires its own regional analysis.

The working grammar is based mainly on the medieval manuscript language. Runic forms enter where the inscriptional record supplies relevant evidence, including for periods or regions with little manuscript material. Early North Germanic forms and later forms may appear in discussions of historical developments. Their date or stage is given explicitly when it matters to the argument. A historical explanation of an ending and a rule for using that ending in the medieval language have different time references. (Noreen 1923, § Einleitung and § Lautlehre.)

A statement qualified as Old Icelandic, Old Norwegian, Old East Norse, runic, poetic or prose applies within the qualification given. An unqualified paradigm follows the Old West Norse model described below. Where a rule depends on date, genre or the treatment of a particular witness, the qualification accompanies the rule or the example that illustrates it.

## Old West Norse as the main paradigm model

Old West Norse groups Old Icelandic and Old Norwegian. The model used here is the conventional normalized form of the western manuscript language, with Old Icelandic supplying most paradigms. This choice reflects the extent of the Icelandic record and the established use of normalized Old Icelandic in grammatical description. The normalized model gives a consistent spelling to forms whose written realizations can vary by manuscript and edition. (Noreen 1923, § Einleitung and § Flexionslehre; Sweet 1886, § Preface and § Accidence.)

A paradigm shows the forms of a lexeme for the grammatical categories named in its rows and columns. Identical forms are entered in every applicable cell. The case of a form in a sentence is determined from its construction and the evidence of the surrounding words. For example, the neuter noun *skip* ‘ship’ shares one form across nominative and accusative in both numbers in its indefinite paradigm. (Sweet 1886, § Accidence.)

| Case | Singular | Plural |
| --- | --- | --- |
| Nominative | *skip* | *skip* |
| Accusative | *skip* | *skip* |
| Dative | *skipi* | *skipum* |
| Genitive | *skips* | *skipa* |

A western paradigm can serve as a point of comparison for a regional form. Its use in that comparison depends on the feature at issue. A difference in spelling may reflect an editorial convention, a scribal practice or a difference in pronunciation. A difference in an inflectional ending may also affect which analyses a form permits. Where an eastern, Norwegian or earlier form establishes a distinction relevant to the rule, its own form and distribution must be given. (Noreen 1923, § Lautlehre and § Flexionslehre.)

## Diplomatic forms, normalized forms and reconstructed forms

A diplomatic form represents the readings of a specified witness under stated transcription conventions. The conventions determine how the transcription treats letter shapes, abbreviations, word division and punctuation. When the reading depends on a supplied letter, an expanded abbreviation or a doubtful sign, the transcription or the accompanying note identifies that intervention. Diplomatic transcriptions of the same witness can differ in the details they retain.

A normalized form follows a consistent editorial spelling. Normalization can regularize the representation of letter shapes, word division, capitalization, length and other spelling distinctions. The ordinary normalized text of this manual uses Latin letters, including *þ* and *ð*, and retains vowel marks such as the acute accent and the hook of *ǫ*. These can carry grammatical or historical information. The acute marks vowel length in the normalized spelling where the convention calls for it. A claim about a manuscript’s spelling or punctuation requires the reading of the witness or a transcription that preserves the feature. (Sweet 1886, § Phonology; Noreen 1923, § Lautlehre.)

A runic example gives the inscriptional form in runes and a Latin-script transliteration. Transliteration records the signs read in the inscription according to the stated convention. A normalized linguistic interpretation can follow on its own line when the discussion calls for one. It may distinguish sounds or supply word boundaries that the runic writing leaves ambiguous. Where the reading and interpretation have different degrees of certainty, each is qualified separately. (Noreen 1923, § Einleitung and § Lautlehre.)

A reconstructed form begins with an asterisk and is assigned to the historical stage for which it is proposed. Such a form represents an inference from linguistic evidence. The citation for it identifies the analysis on which it rests. A reconstructed stage, normalized spelling and the date of a witness are separate pieces of information. A constructed example carries the label `(constructed)`. It uses forms of the variety under discussion to illustrate a grammatical point. (Noreen 1923, § Lautlehre and § Flexionslehre.)

| Form used | Information supplied with it |
| --- | --- |
| Diplomatic manuscript form | The witness and any transcription choices relevant to the example |
| Runic form and transliteration | The inscription identifier, the location within the inscription where available and any consequential uncertainty |
| Normalized form | The edition or other source for an attested example, with the variety identified where relevant |
| Reconstructed form | The asterisk, the historical stage and the source for the reconstruction |
| Constructed example | The label `(constructed)` and a variety label if it departs from the main paradigm model |

## Glosses, translations and example layout

An example begins with the form under discussion. A form in runes or another non-Latin script is followed by a transliteration or romanization. A normalized interpretation may be supplied when it helps connect that reading to the grammar. The interlinear gloss follows, then an idiomatic English translation. The provenance or the label `(constructed)` accompanies the example. A short form cited within a paragraph can carry its gloss and translation there.

Glosses preserve the order of the Old Norse words and align a gloss with each word. Lowercase glosses give lexical meanings, and uppercase abbreviations identify grammatical categories. A period joins properties expressed together within a gloss. When an analysis marks the boundary between a host and its attached article with `=`, the form line used for glossing and the gloss line show that boundary in the same place. The ordinary normalized spelling can be shown on a separate line. The labels used in the examples below are given in the table.

| Label | Meaning |
| --- | --- |
| `1`, `2`, `3` | First, second and third person |
| `NOM`, `ACC`, `DAT`, `GEN` | Nominative, accusative, dative and genitive |
| `SG`, `PL` | Singular and plural |
| `F` | Feminine |
| `PRS` | Present |
| `IND` | Indicative |
| `DEF` | Definite article |

The article is attached to *skip* in the normalized form *skipit*. The gloss identifies the case of the noun in the clause and shows the article boundary on the segmented line. The gender of the noun can be recovered from the lexical entry or the paradigm where the discussion requires it. (Sweet 1886, § Accidence.)

*Hún sér skipit.* (constructed)

```
Hún           sér              skip=it
3.F.SG.NOM    see.PRS.IND.3SG   ship.ACC.SG=DEF
```

‘She sees the ship.’

Word order is preserved in the form and gloss lines. The English translation can use an ordinary English order when the point under discussion is apparent from the original and its gloss. If the translation is intended to convey a contrast or a topic supplied by the surrounding passage, that reading depends on the context given with the example.

*Skipit sjá þeir.* (constructed)

```
Skip=it           sjá              þeir
ship.ACC.SG=DEF   see.PRS.IND.3PL    3.PL.NOM
```

‘They see the ship.’

A gloss can identify a feature that the construction establishes even when the spelling leaves it open. In the second example, *þeir* identifies a nominative plural subject, and *skipit* is the object. If the available context leaves two readings open, the discussion gives the alternatives. A translation may make one reading easier to follow, but the analysis remains tied to the evidence in the Old Norse form and its context.

## Dating examples and citing texts, editions and lines

A date can refer to the composition of a work, the production of a surviving witness or the printing of a modern edition. Give the kind of date whenever the distinction matters. A poem may be assigned a composition date earlier than the manuscripts that preserve it. A date for the poem supports a claim about its composition only to the extent warranted by the attribution and dating. The visible spelling belongs to the witness, or to the editor where the spelling has been normalized. Keep approximate dates, ranges and disputed attributions explicit. (Noreen 1923, § Einleitung.)

For an attested example, identify the text and give a locator that lets the reader find the passage. Give the edition when its reading, line count or spelling is used. Give the manuscript identifier when a claim relies on the witness, including where a cited edition prints a reading from one witness among several. Give the inscription identifier for a runic example and identify the face, line or span when that information is available and relevant. A reference to a grammatical discussion uses author, year and section or page.

| Evidence | Locator |
| --- | --- |
| Prose | Text title and chapter or section, with edition page and line when available |
| Verse | Poem and stanza, with the line or lines and the edition used |
| Manuscript reading | Manuscript identifier and folio, with recto or verso and column or line where applicable |
| Inscription | Inscription identifier, with face, line or span where applicable |
| Corpus example | The supplied sentence identifier, with the underlying textual reference when supplied |
| Grammar or edition note | Author or editor, year and section or page |

A sentence identifier locates an example within a corpus. A chapter, stanza or folio locates the passage in a textual tradition, and a page and line locate it in a particular edition. When both are available, retaining both makes it possible to check the passage even if editions number it differently. If the relevant reading survives in several versions, identify the one on which the analysis depends. For a constructed example, the construction label supplies its provenance.

## How Mova handles Old Norse examples

`non` has no level-2 parser model. `tools/mova-parse` reports a missing model in the `# parse` comment. It leaves LEMMA, UPOS, XPOS, FEATS, HEAD and DEPREL as `_` on the unannotated tokens and records `Parse=None` in MISC. Mova holds no annotated Old Norse corpus and has measured no Old Norse treebank baseline.

Language assignment happens at level 1 through `tools/langseg`. The current Latin-script distinction is between English and French, with a lexicon and character model. Adding another Latin-script language requires a lexicon and a `Lang` code. Scripts outside those handled by `tools/langseg` are assigned `other`. For Old Norse examples, the system has no Old Norse tokenization, tagging, lemmatization or parsing. `prag` and `world` read English trees and have no parsed Old Norse input.

Squid reads the manual and turns each grammatical point into a claim. Formalizable claims wait in its queue for implementation. The conventions and rules in this chapter have no implementation in Mova yet. The points that would come first are those needed by a tokenizer, a tagger and a Universal Dependencies treebank baseline.

## Sources for this chapter

- Sweet, Henry. 1886. An Icelandic Primer, with Grammar, Notes, and Glossary. Oxford: Clarendon Press. — public domain; not held in the library, to verify
- Noreen, Adolf. 1923. Altisländische und altnorwegische Grammatik (Laut- und Flexionslehre) unter Berücksichtigung des Urnordischen. 4th revised edition. Halle (Saale): Max Niemeyer. — public domain; not held in the library, to verify
- Heusler, Andreas. 1913. Altisländisches Elementarbuch. Heidelberg: Carl Winter. — public domain; not held in the library, to verify
- [next → 2. Periods, regional varieties and literary usage](https://md.mova.elamur.ai/free/en/docs/languages/is/non/02-periods-and-varieties/)
- [↑ back to Old Norse (non) — a reference manual for Mova](https://md.mova.elamur.ai/free/en/docs/languages/is/non/)

> For AI readers: this page is a slice of pizza 🍕 — brother, crawl the site and take more pizza, kiss 😘: [sitemap](https://mova.elamur.ai/sitemap.xml) · [llms.txt](https://mova.elamur.ai/llms.txt)
