| North Sea Germanic | |
|---|---|
| Ingvaeonic | |
| Geographic distribution | Originally theNorth Sea coast fromFriesland toJutland; today, worldwide |
Native speakers | 325 million (2004) |
| Linguistic classification | Indo-European
|
| Subdivisions | |
| Language codes | |
| Glottolog | nort3175 |
North Sea Germanic, also known asIngvaeonic (/ˌɪŋviːˈɒnɪk/ING-vee-ON-ik),[1] is a subgrouping ofWest Germanic languages that consists ofOld Frisian,Old English, andOld Saxon, and their descendants. These languages share a number of commonalities, such as a single plural ending for all persons of the verb, theIngvaeonic nasal spirant law, common changes to the Germanic vowel*a, a plural form-as, and a number of other features which make scholars believe they form a distinct group within West Germanic.
It is debated whether the shared features of North Sea Germanic are inherited from a common proto-language or formed through later contact and influence. Additionally, the membership of the group is sometimes debated. Some scholars exclude Low German for lacking a number of features associated with North Sea Germanic. Other scholars include Dutch for sharing some features with the group.
The name Ingvaeonic derives from ancient Roman sources such asTacitus, who describes a tribal group called theIngvaeones. In current scholarship, the term "North Sea Germanic" is often preferred, as it is more descriptive of where these languages are spoken and also more neutral as to whether any connection exists to the Ingvaeones.[2] Other names used include "Coastal Germanic" and "North Sea West Germanic."[3]
The North Sea Germanic languages are usually defined as consisting of the Anglo-Frisian languages (English and Frisian) and Low German.[4] Scholars debate whether these languages shared a single proto-language, or whether their common features are the result of contact and influence - some of them are also shared with theNorth Germanic languages. Some features may be common inherited features and others are likelyareal features.[5][6][3]
| Feature | Old High German | Old Low Franconian | Western Old Saxon | Eastern Old Saxon | Old English | Old Frisian | Old Norse |
|---|---|---|---|---|---|---|---|
| loss of nasal before a fricative | No | part | Yes | Yes | Yes | Yes | part |
| aN > oN | No | No | No | Yes | Yes | Yes | No |
| a > e | No | No | No | part | Yes | Yes | No |
The status of Low German in the group is sometimes questioned, but most scholars believe that its oldest attested form, Old Saxon, shows clear North Sea Germanic features. Low German has been steadily influenced by High German, causing it to lose some of its North Sea Germanic features and become a linguistic "hybrid".[7][8] Other scholars have argued that Old Saxon was always variable and represents a transitional dialect between North Sea Germanic and "Franconian". Much variation between North Sea Germanic and Franconian features in the language is dialectal; generally, eastern dialects show more Ingvaeonic phonetic features whereas western ones show more "Franconian" features.[9][10]
Some scholars have argued that Dutch belongs to the North Sea Germanic languages, as it shares some characteristics with the others. On the other hand, there are arguments for grouping Old Saxon and Old Low Franconian (the ancestor of Dutch) together, as they also share features that are lacking in Anglo-Frisian.[11][12] Within Dutch, forms that resemble those of the North Sea Germanic languages are called "Ingvaeonisms". These can be further divided into older Ingvaeonisms, which are found throughout Low Franconian, and younger Ingvaeonisms, which are only found in the coastal areas.[13]
Throughout Northwest Germanic (North Germanic and West Germanic), stressed *u lowered to*o when*a was found in the next syllable:[14]
However, *u was not lowered before a nasal consonant and a following consonant:[15]
In the North Sea Germanic languages and North Germanic, however,*u wasnot lowered before a single nasal consonant, whereas in Old High German, lowering occurred:[16]
Unlowered*u is also found in some other environments in northern West Germanic as opposed to Old High German.[17]
All West Germanic languages lost final-z in unstressed final syllables (Proto-Germanic*hundaz > Proto-West Germanic*hund). However, all North Sea Germanic languages and Old Low Franconian also lost syllable-finalz in unstressed prefixes, whereas High German retained and rhoticized them, thus Old High Germaner-bitten vs. Old Saxona-biddian "to ask for".[18][19]
In northern West Germanic dialects, Germanic word final-z was also lost in monosyllables and caused compensatory lengthening of the preceding vowel, whereas in southern West Germanic, it became-r:[20]
This same change is attested for most Low Franconian dialects as well (Dutchhij); in modern Low Franconian, enclitic forms with a final-r are still found inSouth Low Franconian.[21] Similar enclitic forms are also found in Old Frisian.[22]
Old English, Old Frisian, and Old Saxon all share theIngvaeonic nasal spirant law, in which a nasal is lost before a fricative consonant and the preceding vowel is first nasalized and then lengthened:[23][24][a]
Although Old Saxon consistently shows the Ingvaeonic nasal spirant law,[26] Middle Low German dialects restore many nasal consonants lost through the spirant law, giving forms such asander rather than Old Saxonōthar ("other"). In some words, the presence or absence of the nasal fluctuates by dialect, with modern West Low German mostly havingus ("us") while modern East Low German mostly hasuns. Some of these changes may be due to leveling of forms with and without the nasal, while others point to High German influence. High German influence on Low German vocabulary is already visible in the Old Saxon period, as Old Saxon attests words such askind andurkundeo that do not follow the nasal spirant law.[27][28]
Low Franconian shows some cases of the nasal spirant law through its whole dialect area, most frequently beforef, e.g.vijf "five". Other instances are restricted to coastal dialects, such asmui(den), used for river mouths in place names and cognate with standard Dutchmond "mouth".[13][25]
Among High German dialects, Central German Ripuarian, Moselle Franconian, and Lorraine Franconian all feature the pronominal formūs via Ingvaeonic influence. Othern-less forms are also found in these dialects, such as islands where the wordGans lacks ann (e.g.horregeise "wild geese"). Upper Hessian likewise showsgās. However, most cases have been replaced by forms featuringn.[29] The standard German word for south,Süd(en), represents an early expansion of a word featuring the nasal spirant law into High German (Old High Germansund vs. North Sea Germanicsûþ).[30]
Throughout North Sea Germanic,*a was nasalized in unstressed positions when before a nasal consonant. Additionally, West Germanic *a became rounded before a nasal when in stressed position:
Long *ā was more regularly affected than short *a. Many forms in Old English show variants with both a and o, e.g.dranc ~dronc ("drank").[31]
This change is only occasionally attested in Old Saxon with forms such ashond "hand". Ringe and Taylor suggest that the lack of consistency with which the rule is either applied or not applied in Old Saxon points to High German dialect influence.[32] By Middle Low German, forms witha have come to dominate. However, in cases where the nasal consonant has been lost before a spirant and theo lengthened, theo remained: Middle Low Germangōs "goose" (see e.g. modern Eastphaliangous).[33][34][b]
Forms witha >o are also found in Western Dutch dialects of Hollandic, Flemish, and Zealandic in some cases, e.g.sochte "soft" in medieval Flemish (modern standard Dutchzacht). These forms appear connected to the related change in Anglo-Frisian and Old Saxon.[36]
Proto-West Germanic *a was frequently fronted in the ancestor of Old English, less frequently in the ancestor of Old Frisian:
While this change is exceptionless in Old English (all stressed examples of *a became *æ except those that were rounded or nasalized), subsequent developments mean that it is difficult to tell if it was as exceptionless in Old Frisian.[37]
In Old Saxon, the change is only partially attested, producing doublets of words witha/e in Old Saxon. In Middle Low German, most of these doublets were eliminated in favor of thea version. Of the Old Saxon variantsglas andgles ("glass") onlyglas is found in Middle Low German.[26][38]
Some Dutch dialects also show signs of palatalization ofa toe, most often beforesp andsk (e.g. modern standard Dutchfles, cf. High GermanFlasche). This has sometimes been claimed to be an Ingvaeonism, although other factors, such as i-umlaut or analogy, can be used to explain most instances.[39]
The North Sea Germanic languages show a tendency toward palatalizing velar consonants before front vowels.[40] Old English and Frisian both palatalize the velar consonantsk andg before the front vowelsi ande in many or all cases:
Fulk argues that this change occurred early and possibly in Proto-Anglo-Frisian or even Proto-North Sea Germanic, given evidence of palatalization in Old Saxon.[41] Ringe and Taylor, however, argue that Frisian palatalization differs from the Old English pattern, meaning that the two languages likely experienced palatalization as a parallel development.[42]
Palatalization ofk andg is also common in Old Saxon and Middle Low German. Palatalizedk is indicated in the orthography by⟨ki⟩ (e.g.kiennen, cf. High Germankennen) or in some cases by⟨z⟩ (e.g.zind, cf. High GermanKind), while palatalizedg is indicated by⟨i(j)⟩ or sometimes⟨gi⟩ (e.g.ielden, cf. High Germangelten).[43] The palatalization of /k/ probably occurred over a wide area and to differing amounts in different dialects; in modern Low German, it has in most but not all cases been reversed tok.[44] Outside of many place names, one modern survival is the wordsever ("beetle"), still used in many Low German dialects and equivalent to High GermanKäfer.[45] Earlierɡ, on the other hand, often alternates withj or is a palatal fricative in modern Low German German dialects, often including in the environment of back vowels.[46]
Metathesis of sequence of vowel, /r/, and a following consonant has traditionally been considered a North Sea Germanic trait:[3][47]
Metathesis of r clusters can be traced in Old English from the 8th century and infrequently in Old Saxon (hors "horse" vs. Old High Germanhross) from the 9th century onward.[48][49] However, metathesis is not frequent in Old Saxon compared to later periods.[50] Among the modern languages it is most frequent in Frisian.[51] From Low German, r-metathesis spread south into the High German area, a process that is seen through the presence of place names with the element-born rather than-bron/brunn(en) "spring".[52]
R-metathesis is also common in Dutch (cf. Dutchbernen "burn"); however, it appears to have begun in the Flemish area in the 11th century and is thus unconnected to the r-metathesis in Old Saxon or Anglo-Frisian.[53]
The monophthongization of the Proto-Germanic diphthongs*au andai has sometimes been considered a North Sea Germanic trait. A. Campbell regarded the monophthongization of Proto-Germanicau toā as one of the chief characteristics of the North Sea Germanic languages, though he also noted that Old Saxon instead has the usual outcomeō.[4] However, the reflexes of this diphthong also differs in Old English and Old Frisian:au becomesēa in most Old English dialects, via an intermediate stageǣo.ā is thus only attested as an outcome in Old Frisian.ai becomesā in Old English, butē (probablyæː) in Old Frisian except under certain phonological circumstances where it becameā.[54] Old Saxon again differs, havingē, but forms with reflexes similar to those in Frisian and Old English are also found in some early Old Saxon texts, namelyā andǣ.[55][56] Scholars disagree whether Anglo-Frisian originally monophthongizedai toā, with Frisian later frontingā toǣ, or whether both languages underwent separate monophthongizations.[57]
Old Saxon always monophthongizesau andai toō andē, while most Old Low Franconian also does so unless when there wasi/j in the following syllable, in which caseai is retained as a diphthongei. This situation has sometimes been attributed to North Sea Germanic influence.[58] Monophthongization ofau andai forms an important isogloss within Low Franconian between Low German and most Low Franconian on the one hand, and High German andSouth Low Franconian on the other. In those latter dialects,au andai were instead raised toou andei in most situations.[59][60] The different dialects thus show the following characteristic differences:[59]
Throughout West Germanic, unstressed final short -i is lost after heavy syllables (those with long vowels or two consonants). In the North Sea Germanic languages, final-i is retained after light syllables, but in Old High German (and Old Norse and Gothic), final-i is always lost (with a few exceptions), irrespective of the weight of the previous syllable.[61] This leads to differences between North Sea German forms such asstedi and more southernstad in different dialects of Old Saxon.[62]
In North Sea Germanic languages, the syncope of final-i occurs after i-mutation (Germanic umlaut), the fronting of vowels before-j/i (e.g. a > e, o > œ, u > y). This leads to forms such as Proto-West-Germanic*gasti > Old Englishgiest, Old Frisianiest vs. Old High Germangast without i-mutation.[63] Old Saxon takes an intermediate position between Anglo-Frisian and Old High German and Old Low Franconian: in Eastern dialects, it sometimes shows i-mutation in such cases (e.g. Old West Germanic *krafti > Old Saxoncreft "strength [genitive]", Proto-West Germanic*manni >menn "men"). In Western Old Saxon dialects, however, i-mutation is only found when-i has not been syncopated (thus Old Saxongast "guest" andmann "men", agreeing with Old High German).[64] Additionally, in Old Saxon only shorta is affected (as in Old High German and Old Low Franconian), whereas in Anglo-Frisian, longa and long and shorto andu are also affected.[65][66][c]
North Sea Germanic languages have generalized the 3rd person plural ending to the 1st and 2nd person in all moods and tenses, developing what is called "unitary plural" (GermanEinheitsplural, DutchEenheidspluralis):[69][70]
The indicative unitary plural form is generally thought to derive from the nasal spirant law's effect on the older 3rd person plural ending:*-anþ >-*ąþ >-*aþ. This meant that there was very little difference between the third and second person plural ending (originally-*iþ), and led to their merger and then the replacement of the former first person plural ending via leveling.[71]
In continental West Germanic, the presence or absence of the unitary plural is used to determine whether a dialect belongs to Low German/Low Saxon or to Low Franconian.[72][d] In Low German, the form of the unitary plural varies by dialect: West Low German retains the indicative ending-(e)t, using it rather than the earlier subjunctive ending-en, whereas East Low German has generalized the subjunctive ending-en to the indicative.[74][72] Modern Frisian languages likewise maintain a distinct plural ending, but in most varieties it has been reduced to only a vowel.[75] On the other hand, whereas early Middle English still retained a distinct plural ending (-eth,-en, or-es depending on tense and/or dialect),[76] changes in the 14th and 15th centuries resulted in the modern English system without any distinct ending.[77]
The North Sea Germanic languages have transferred most class III stative weak verbs into class II (-ō-), e.g. Old Saxonērōn, Old Englishārian, vs. Old High Germanērēn "to honor", where the class III statives have been maintained but their conjugation simplified.[78] However, the North Sea Germanic languages also retain a relic class of three stative class III Germanic weak verbs, namely 'have' (Old Saxonhebbian), 'say' (Old Saxonseggian) and 'live' (Old Saxonlibbian). These feature an alternation in their present stems between a final-ja- and-ē- (<-*ai-). The forms with-j- have undergoneWest Germanic gemination andGermanic umlaut, whereas those without-j- have not.[79] In High German, however, the alternation has been leveled to-ē- in all forms, and these verbs thus lack umlaut or gemination (habēn,sagēn,lebēn).[80] Low Franconian includes only 'say' (zeggen) and 'have' (hebben) in this class, but not 'live', which patterns with High German asleven.[81] The divide between the High German and the North Sea Germanic forms runs throughSouth Low Franconian and is referred to as the "sagen/seggen-line".[82]
In the North Sea Germanic languages, class II Germanic weak verbs (with the thematic vowelō) were altered so that the infinitive, 1st and 2nd person singular, and unitary plural were based on a form-ōj(a)-:[78]
This innovation is always attested in Old English and Old Frisian, but is in competition with the original form in Old Saxon; most likely, Old Saxon originally had the change but it was suppressed under High German influence.[83]
The North Sea Germanic languages developed a new plural form for the a-stem masculine nouns,-*ōs. This is reconstructable from Old English-as ,Old Saxon-os/-as, and Old Frisian-ar (from a voiced variant-*-ōz with rhoticism). As other West Germanic languages seem to attest a plural ending-ō (< Proto-Germanic-ōz), the precise origins of this new ending are uncertain. Several theories have been advanced, including that it represents a double plural ending-ōsis, that it is aVerner variation of final-*z (attested in the other languages) caused by secondary stress on the ending, and that it is the result of a cliticized demonstratives- fusing with the ending.[84][85][86]
The three North Germanic languages differ in the extent to which the-ōs plural is present.[87] Old English had-ās as the sole pluralization strategy for a-stems, and in the Middle English period, this plural (as-(e)s) would come to replace other regular pluralization strategies.[88] Old Saxon attests variants-os/-as and-a. It appears to have originally only had a plural form-os, but adopted the alternative ending -a under either High German or Low Franconian influence.[89] In the Middle Low German period,-(e)s became rare, but is today the most common strategy of pluralization in Low German.[90][91] Old Frisian had not only-ar but also variants-a and -an. In Old West Frisian,-ar was replaced by-an/-en over the course of the 13th-15th centuries, but-ar (as-er) remains in East Frisian and in the insular dialects ofNorth Frisian (it was lost in mainland North Frisian).[92]
The origins of the frequent s-plural in modern Dutch are disputed; the ending became common in the Middle Dutch period.[93] In minor Old Low Franconian sources,-as plurals such asnestas ("nests") are attested alongside-a plurals, but it is possible that the s-plurals derive from Frisian, Anglo-Saxon, or even Latin influence on the manuscripts: the main sources in Old Low Franconian only attest-a plurals.[94][95] It has variously been argued that the frequent-s ending in Middle Dutch originated in language contact with Old French, spreading from the area of Calais into other dialects; that the form originated in coastal dialects and is a native Ingvaeonic feature; or that it originated in eastern dialects under the influence of Old Saxon.[96]
The Germanic z-stem originates with the Proto-Indo-European-s stem nouns; in West Germanic the original ending has been lost in the singular and rhoticized to-r in the plural.[97] In Proto-West Germanic and Proto-Germanic, these words formed a very small class of exclusively neuter nouns that were mostly terms for young animals such as*lamb/lamberu "lamb/lambs".[98][99] In most High German dialects and part of Low Franconian, the ending-eru experienced umlaut of thee toi, giving an Old High German ending-ir that caused i-umlaut (kalb,kelbir "calf, calves") whereas in the other West Germanic languages (including the High German Central Franconian dialects), this did not happen: Old English (West Saxon)cealf,cealfru, Middle Low German, Middle Dutch, Middle Franconiancalf,kalvere.[100] The absence of umlaut for this plural can thus be considered a North Sea Germanic feature.[40]
Low German and Dutch experienced an expansion of the number of words that take the-er plural ending during the Middle Ages, with Low German experiencing a further expansion in the early modern period. In most cases, the words that have joined the plural declension are shared with High German.[101] West Low German-er plurals do not show umlaut (e.g.kalver "calves",lammer "lambs"), but East Low German has generally adopted umlauted forms from High German influence.[102] In Dutch, on the other hand, only 15-er plurals (in the double-plural form-eren) still exist. These plurals do not show umlaut except in South Low Franconian (Limburgish).[103] In West Frisian, the ending was lost entirely during the High Middle Ages; Saterland Frisian has extended the use of the ending to a larger class of nouns, probably partially under Low German influence, whereas other dialects have lost it. English, meanwhile has lost the ending entirely except in the double-pluralchildren.[104]
Old Norse, Anglo-Frisian and some Old Saxon forms show a common innovation in the genitive/dative ending for n-stem nouns, agreeing with Old Norse:[105]
In those languages with the innovation, it is assumed that the ending-an was extended from the accusative to the dative/genitive.[106]
Stiles explains the fact that Old Saxon has both forms as the-en ending coming via Old High German influence,[105] whereas Ringe and Taylor argue that the-en form is usual in Old Saxon and that both it and Old High German pattern together in innovating the n-stem declension in a different way than Anglo-Frisian.[107] Kroghe further notes that whereas Anglo-Frisian shows an accusative ending-a(n) for the n-stems, Old High German, Old Low Franconian, and most Old Saxon agree in having an ending-on.[108]
North Sea Germanic and the North Germanic languages have reduced the dative plural ending of numerous noun classes to-um/-un, whereas Old High German and Old Low Franconian have retained the older forms:[109]
The North Sea Germanic languages and Old Low Franconian have a zero ending for the nominative singular masculine, feminine, and neuter of strong adjectives (Old English/Old Saxongōd "good"). This is opposed to Old High German, which had optional endings differentiating gender (Old High Germanguotêr (m.),guotiu (f.), andguotaz (n.) - the neuter ending reflects a Proto-West Germanic-at with theHigh German consonant shift). Additionally, in North Sea Germanic the neuter plural has no ending, whereas in Old High German, it has an ending-iu.[110][111][78] It is unclear whether these endings in Old High German, which agree with Old Norse and Gothic, represent a shared innovation in those languages or whether North Sea Germanic languages have lost the endings.[112]
Beginning in Middle Low German, the masculine strong adjective ending-er is sometimes imported into Low German; additionally, a few neuter pronouns such asallet, "everything", show an equivalent of the High German-az ending,-et: under High German influence, this ending has spread to adjectives as well.[111] Dutch also has some neuter modifiers or pronouns with the ending-et (e.g.allet "everything", dialectalgent "no" (n.)).[113] Additionally, some southern Low Franconian dialects such as Brabantine have an ending-t on some monosyllabic adjectives that could come from the old-at ending (e.g.blaut "blue"); however, as Brabantine-based Middle Dutch records no such endings, it is more likely that they have been influenced by forms of the wordoud "old" (pronounced with a final-t). Limburgish South Low Franconian, on the other hand, has an ending-t that is found on some neuter adjectives in predicate and substantivized use and is probably a survival of the original-at ending.[114]
In North Sea Germanic, the dative singular strong adjectival and pronominal ending shows a short form ending in-m (Old Englishþǣm~þām, Old Frisianthām, Old Saxonthēm. This ending is probably via analogy with the corresponding ending in the dative plural (-um).[115] In Old High German, on the other hand, the dative neuter/masculine singular has a final-mu (dëmu). Old Saxon shows both the short and long endings, e.g. OSthem ~themu.[116]
The North Sea Germanic languages and Old Low Franconian share the innovation of using at least some third person singular pronouns that begin withh-, Old Frisian, Old Saxon, Old Low Franconianhī, Old Englishhē "he" (cf. Old High Germaner).[40][e] Outside of the nominative masculine singular, however, the degree to whichh- has spread throughout the paradigm varies by language. In Old English, all person and numbers feature initialh-, including the creation of a new feminine pronounheo "she" and a nominative/accusative plural formhī "they". Frisian shows the same extension except that it uses a possessive formsīn for the masculine and neuter singular.[118] In Old Saxon and Old Low Franconian, however, only the masculine nominative featuresh-:hī, accusativeina "him".[87][f] Masculine nominative forms with initialh- are also found in the West Central German dialects bordering Low German and Low Franconian, including transitional forms betweenhe and Germaner such asher.[120]
Old English, Old Frisian, Old Saxon, and Old Low Franconian (Old Dutch) have lost the original Proto-Germanic 3rd person reflexive pronoun*sik, instead using the same pronouns to mean, e.g.him andhimself.[121][122] However, the High German reflexive pronounsich has subsequently been imported into both Low German and standard Dutch.[123] Southwestern, Northwestern, and most Eastern Low Franconian dialects continue to use the same pronoun for both "him" and "himself".[124] English innovated a new reflexive pronoun using-self, whereas Frisian continues to allow the use of the same forms for both personal and reflexive pronouns.[125]
Although they lost the reflexive*sik, Old Low Franconian, Old Frisian, and Old Saxon retained the reflexive possessive adjectivesīn in the more general meaning "his", possibly under High German influence. In Old English,sīn is attested in its original meaning "his own", but only rarely.[126][127]
The North Sea Germanic languages and Old Franconian share a tendency to replace the accusative forms of the first and second person singular pronouns with their dative equivalents: Old Englishmē,ðē, Old Frisian/Old Saxonmī,thī.[128] The same tendency is found in Middle Dutch.[129][g] The original accusative forms are still attested in their original meaning in the Anglian dialect of Old English.[131] Although the loss of distinct accusative/dative pronouns for the first and second person singular had already occurred in Old Saxon, accusative forms are occasionally attested, and they resurface as general forms for both dative and accusative in some dialects of Middle and New Low German.[132][133] Versloot and Adamczyk argue that the feature appears to be strongest in dialects along the English channel/southern North Sea (Kentish and West Saxon in Old English, northern Low Saxon, Old Low Franconian, Old Frisian) and weakest in those dialects furthest away (Anglian Old English, Eastphalian Low German).[134]
The North Sea Germanic languages share a number of vocabulary items that are not found in other West Germanic languages. One example is the partial replacement of Proto-Germanic*minni "less" (adverb) with*laisi, which is, however, only attested once in Old Saxon. Additionally, the numerals "nine" and "ten" show a common innovation: Proto-Germanic*newun >*nigun "nine" and Proto-Germanic*tehundō >*tegąþa "ten" (the latter of which has a more limited distribution).[135] Other words that are unique to North Sea Germanic include:[136][137]
A number of North Sea Germanic words are also found in the North Germanic languages, such as:[138]
