Explanation of Summary Categories

The Summary Category provides a shorthand reference indicating both the nature of the etymological evidence and our assessment of the strength of each individual case for Old Norse input. This follows the categorisation system developed by Richard Dance (fully detailed in Dance; see bibliography for details). 

Four types
There are four orders of evidence, represented by capital A-D, with each successive letter indicating the broadly increasing level of uncertainty involved with the structural type of evidence in question. Within each type, however, the relative quality of the arguments for Old Norse input varies for individual lexemes (thus not all type Bs are inherently more likely to derive from Old Norse than all type Cs or Ds, etc.). For this reason decreasing plausibility within types is also indicated, with a doubling or tripling of the initial capital. This does not apply to type A, which is by definition most secure. 

Type A indicates systematic, formal evidence and thus the strongest case for Old Norse input (although it should be stressed that the identification is only as secure as the postulated pre-history of the word). There are three subtypes:

A1: secure phonological evidence (e.g. Middle English blayke ‘yellow’ shows Old Norse /ei/ < Proto-Germanic /ai/; cp. Old Icelandic bleikr ‘pale, wan, ?bleached’ and Old English blāc ‘bright, shining, glittering; pale, pallid, wan’). Where phonological evidence is not secure, the lexeme is assigned to another type.

A2: morphological evidence (e.g. Middle English haʒer (MED hauer) shows Old Norse inflexional (nominative, masculine adjectival) -r; cp. Old Icelandic hag-r ‘handy, skilful’ and Old English hæg-(hāl) 'healthy')

A3: phonological AND morphological evidence (e.g. ME tite with ON adjectival (adverbial) -t AND ON consonant cluster assimilation; cp. Old Icelandic títt < *tíð-t ‘quickly’ and Old English tīd (n.) ‘time’)

Additionally, an asterisk (*) appended to any of the above indicates the survival of a native English cognate for comparison. Thus we have six possible types in this category: A1, A1*, A2, A2*, A3 and A3*.

Type B indicates lexemes formed on Germanic roots which are not attested early enough in Old English (i.e. before Anglo-Norse contact). There are two subtypes: 

B1: the root is only otherwise known in North Germanic (e.g. Middle English meke ‘gentle, submissive, merciful, humble’ from a Proto-Germanic root *meuk- which is only found in this form in Scandinavia; cp. Old Icelandic mjúkr ‘soft, agile; meek, mild’), and thus it is possible to argue that the likelihood of Old English derivation is more likely than with B1.

B2: the root is attested in at least one other Germanic language (beyond North Germanic) (e.g. Middle English take ‘to take, etc.’ is a strong verb formed on Proto-Germanic *tak-; cp. Old Icelandic taka ‘to take’ in contrast to the weak formations attested in Middle Low German and Middle Dutch) 

Type C indicates that the root is known in early Old English (or there is an unambiguous form-source in a third language), but some aspect of form, sense or usage suggests Scandinavian influence. There are five subtypes, indicating the main feature of the Middle English word for which Old Norse input has been sought as an explanation:

C1: derivational form (e.g. Middle English anger ‘harm, sorrow; anger’ with -r from an original neuter es-stem, only attested in Old Norse, although the root itself is also found in Old English; cp. Old Icelandic angr ‘sorrow, resentment, distress; repentance; tribulation, injury’ and Old English ange (adj.) ‘anxious, painful, distressing’)

C2: inconclusive orthographic or phonological form (e.g. Middle English forms of the verb bresten ‘to break; burst’ without the metathesis otherwise apparent in West Germanic; cp. Old Icelandic bresta ‘crack, crash; snap, break, be split open’ and Old English berstan ‘to break, burst, fail, fall’)

C3: meaning (e.g. Middle English ande has the sense ‘breath’ which is not attested in Old English; cp Old Icelandic andi ‘breath, breathing, spirit’ and Old English anda ‘malice, envy, hatred, zeal’)

C4: formation of a compound or phrase (e.g. Middle English omys (adv.) ‘amiss’; cp. the Old Icelandic phrase (á) miss ‘so as to miss’ and Old English mis- (prefix) and missan (v.) ‘to miss, escape the notice of’)

C5: frequency (e.g. Middle English til ‘to; until’; cp. Old Icelandic til ‘to, until, etc.’ and (rare Northumbrian) Old English til ‘to, until’ alongside much more frequent tō, etc.)

Secondary labels and prefixes
Multiple criteria may apply to this type, in which case a secondary category is also given in the database entry. 

When a word’s form cannot derive from Old English and belongs (ultimately or more recently) to a non-Germanic language which could have been its direct source, the category label is prefixed by F. Thus Middle English cross is categorised as FC5b. Late Old English cros ultimately derives from Latin crux ‘cross’ (oblique stem cruc-), but the proximate source is debated, with possibilities including Olr cros, ON (cp. OIcel kross), Anglo-French croscecross and medieval Latin cros.

Decreasing probability for types B and C are noted as:

BB or CC: where an alternative explanation is equally plausible and scholarship is divided

BBB or CCC: the case is weak or the word’s history can be otherwise explained, but it is not possible to rule out Old Norse input categorically

Type D indicates an uncertain or contested source form. There are two subtypes:

D1: form and sense can be established, but there is no generally accepted etymology (e.g. Middle English lasse ‘girl’ of obscure etymology, possibly connected with early Swedish compounds loska-mader ‘bachelor’ and loska-kona ‘unmarried woman’) 

D2: the interpretation of the word itself is contested in its Middle English textual context (e.g. Middle English breth at Cleanness 916 has been read as ‘breath’, continuing Old English brǣð ‘breath, odour, exhalation, etc.’ or ‘fury’, in which case cp. Old Icelandic bræði ‘anger, fury’)

In the case of type D, there are almost always plausible alternative explanations to Old Norse influence, and so only one degree of decreasing probability is noted:

DD: the case is weak or the word’s history can be better explained otherwise, but it is not possible to rule out Old Norse input categorically


Circumstantial Evidence

In addition to the four structural types of evidence denoted by the initial capitals detailed above, the summary categories also indicate the presence of additional circumstantial evidence with lowercase letters a-d appended to the main category label.

a: a continental West Germanic cognate with substantially the same form, sense or usage is attested. This label is applied where that form, sense or usage is the reason for entertaining Old Norse input.

b: principally confined to the North or East in the toponymic record

c: principally confined to the North or East in the lexical record

d: early occurrences in English are strongly associated with Scandinavian cultural influence