The linguistic orthodoxy that sees
Old English as a direct descendant of Old Frisian and later dialects of English as descendants of Old English of the Angles and Saxons that arrived in Britain in the 5th century C.E., even as new loan words were incorporated from Old Norse and following the Norman Conquest in 1066 CE, French and Latin.
But, two academic linguists, Jan Terje Faarlund, professor of linguistics at the University of Oslo and Joseph Emmonds, visiting professor from Palacký University in the Czech
Republic disagree.
They are now claiming that Middle English (traditionally designed as the dialect of English spoken in Britain after the Norman Conquest in 1066 CE) and subsequent dialects were descendants of the
Old Norse language, and that this language replaced the Old English language that arrived in the 5th century CE in the roughly two and a half centuries before the Norman Conquest in 1066 CE. Old Norse is more similar to Middle English and subsequent languages grammatically, even though Middle English had heavy lexical borrowing (i.e. lots of loan words) from Old English during this transitional period.

"Modern English is a direct descendant of the language of Scandinavians who
settled in the British Isles in the course of many centuries, before the
French-speaking Normans conquered the country in 1066," says Faarlund. He points
out that Old English and Modern English are two very different languages.
Why?
"We believe it is because Old English quite simply died out while
Scandinavian survived, albeit strongly influenced of course by Old English," he
says.
The 'cohabitation' between the British and the Scandinavians was largely
hostile. Both fought for political hegemony. The descendants of the Vikings
gained control of the eastern and northern parts of the country. The Danelaw was
under the control of Scandinavian chiefs for half a century [ed. according to Wikipedia Danish mass migration became around 880 CE, Danelaw proper was in place from 886 CE to 954 CE, and this followed by rule by Scandinavian monarch again from 1016 CE until 1044 CE when Edward the Confessor returned the throne to non-Scandinavian rule until the Normans defeated him.]
Like most colonists, the Scandinavian-speaking inhabitants found no
reason to switch to the language of the country they had arrived in.
"One
especially important, geographic point in our study is that the East Midlands
region, where the spoken language later developed into Modern English, coincides
almost exactly with the densely populated, southern part of the Danelaw," says
the professor.
The language adopted many words from the Danelaw's inhabitants who were of
Norwegian and Danish descent. For example, all the lexical words in this
sentence are Scandinavian: He took the knife and cut the steak. Only
he, the and and come from Old English.
"What is particularly interesting is that Old English adopted words for
day-to-day things that were already in the language. Usually one borrows words
and concepts for new things. In English almost the reverse is true – the
day-to-day words are Scandinavian, and there are many of them," says
Faarlund.
Here are some examples: anger, awe, bag, band, big, birth, both, bull,
cake, call, cast, cosy, cross, die, dirt, dream, egg, fellow, flat, gain, get,
gift, give, guess, guest, hug, husband, ill, kid, law, leg, lift, likely, link,
loan, loose, low, mistake, odd, race, raise, root, rotten, same, seat, seem,
sister, skill, skin, skirt, sky, steak, though, thrive, Thursday, tight, till,
trust, ugly, want, weak, window, wing, wrong.
The researchers believe that Old English already had 90 per cent of these
concepts in its own vocabulary.
But the Scandinavian element was not limited to the vocabulary, which is
normal when languages come into contact with each other. Even though a massive
number of new words are on their way into a language, it nevertheless retains
its own grammar. This is almost a universal law.
"But in England grammatical
words and morphemes - in other words the smallest abstract, meaningful
linguistic unit - were also adopted from Scandinavian and survive in English to
this day."
The two researchers show that the sentence structure in Middle English - and
thus also Modern English - is Scandinavian and not Western Germanic.
"It is
highly irregular to borrow the syntax and structure from one language and use it
in another language. In our days the Norwegians are borrowing words from
English, and many people are concerned about this. However, the Norwegian word
structure is totally unaffected by English. It remains the same. The same goes
for the structure in English: it is virtually unaffected by Old English." . . .
"We can show that wherever English differs syntactically from the other Western Germanic languages -- German, Dutch, Frisian -- it has the same structure as the Scandinavian languages."
From
here.
This hypothesis suggests that the transition from "early Old English" to "late Old English" ca. 900 CE, rather than Norman Conquest, really marks the transition from Old English, an Anglo-Saxon West Germanic language, to Middle English, a North Germanic Scandinavian language.
In their theory, Anglo-Saxon derived West Germanic Old English endured for about 450 years rather than 700 years. As the Wikipedia article on Old English notes, the preceding Celtic language of the region was more or less completely displaced in a process that started with Old English.
Traditionally, and following the Anglo-Saxon preference prevalent in the nineteenth century, many maintain that the influence of Brythonic Celtic on English has been small, citing the small number of Celtic loanwords taken into the language. The number of Celtic loanwords is of a lower order than either Latin or Scandinavian.
The displacement of the Celtic languages that were dominant for a far longer period of time prior to the arrival of Old English, than Old English had been in use as of 800 CE, may have cleared the way for an easier language shift to another language that was still in the Germanic language family four and a half centuries later.
Another reason to suspect language shift from Anglo-Saxon Old English to an Old Norse derived Middle English is that many of the toponymn in the region are Norse rather than Anglo-Saxon. Toponyns are often thought to be among the most resiliant evidence of a language that would not replace older usages unless the language shift was particularly complete. Toponymns are like words relavant to words in daily usage for existing concepts that are less prone to be borrowed than are word for newly acquired ideas expressed in a language (e.g. words for imported products or technologies).
The authors argue that this is one of the reason that Scandinavians have such an easy time learning to speak English as a second language relative to speakers of other languages.
For what it is worth, I find their proposal, despite the fact that it contradicts long time linguistic orthodoxy, to be very convincing both as a matter of linguistics and as a fit to a historically documented narrative.
As a native speaker of English with roughly equal parts German and Swedish speaking Finnish descent who is aware of relatives living in both places, who has virtually no ability to speak or write or read either language, and as someone who has read Old and Middle English works and is familiar with the history of the period from his education, I am arguably in a position to be a relatively neutral evaluator of this claim (free of nationalistic bias).
In the larger scheme of language evolution, this thesis is yet another data point to suggest that something like Newton's second law of motion (i.e. inertia) applies to languages as well. Rather than changing over time mostly due to random linguistic drift within a particular culture, a far larger share of language change than has historically been appreciated happens for specific historical reasons involving colliding cultures. Transitions like the transition between Old English and Middle English happen not simply due to the passage of time, but because a distinct superstrate culture imposed new linguistic standards on the general population.
This analysis also illustrates the point that it pays to be skeptical of extremely deep cultural legacies. Britain received a major cultural reboot from the Norwegians just 1200 years ago that wiped away much of its cultural legacy from the early Middle Ages, Roman Period, iron age, bronze age, and earlier Neolithic eras.
This linguistic claim also implies a larger cultural claim. The cultural legacy of the English people may be more Scandinavian than German and Dutch, and the strong Anglo-Saxon cultural influences on English culture (relative to Scandinavian influences) may be an ahistorical myth.
In the context of American culture, the Yankee culture sourced to the English Midlands according to Fischer's Albion's Seed and adopted regionally by Scandinavian immigrants may have in fact itself have been the most Scandinavian of English regional cultures to make it to the new world in the first place. Hence, similarity due to common cultural origins between English and Scandinavian immigrants, rather than similarity due to transmission of local regional culture to new immigrants, may be important in the cultural formation of these parts of the United States' cultural heritage.
Notably, many "grammar myths" which are commonly viewed as prescriptive rules of formal modern English grammar, but are not in fact observed in literature and other writing and speaking by educated native English language speakers in formal settings, involve situations where Norweigan and English grammar differ from West Germanic and Latin grammatical rules.
The university's press release did not reference a new publication by these linguists that has made this case.
UPDATE November 29, 2012:
The researchers conclusions don't change the mainstream classification of English as a Germanic language. They merely reassigns English from one of the two surviving subfamilies of the Germanic languages to the other one.
The linguistic structure of the Germanic languages is outlined below for context. I also distinguish some neighboring non-Germanic language and describe their place in the overall classification scheme for languages.
Old Norse
As Maju notes in the comments and as can be discerned from the link in the text above, "Old Norse" is the language ancestral to Icelandic, Faroese, Norwegian, Danish and Swedish, i.e. to all Northern Germanic languages. It also influenced many other languages and is ancestral to both Middle English and modern English if the researchers discussed above are correct. Modern Icelandic is the modern language that has changed the least from Old Norse in the last thousand years. As Wikipedia explains in its article on Old Norse:
Proto-Norse developed into Old Norse by the 8th century, and Old Norse began to develop into the modern North Germanic languages in the mid- to late 14th century, ending the language phase known as Old Norse. These dates, however, are not absolute, since written Old Norse is found well into the 15th century
The 12th century Icelandic Gray Goose Laws state that Swedes, Norwegians, Icelanders and Danes spoke the same language, dǫnsk tunga ("Danish tongue". . .). Today Old Norse has developed into the modern North Germanic languages (Icelandic, Faroese, Norwegian, Danish and Swedish), and although distinct languages there is still considerable mutual intelligibility.
West Germanic Languages
All of the other living
Germanic languages belong to the West Germanic language family, the most notable representatives of which are German, Dutch, Frisian, Luxembourgish and Pennsylvania German (spoken by the Amish), Yiddish and Afrikaans. The Angles and Saxons who invaded England in the 5th century were speakers of a West Germanic language which is most similar to the modern Friscian language.
East Germanic Languages
There was once an East Germanic language family, but all of the languages in that language family are now extinct. "The East Germanic languages were marginalized from the end of the
Migration period [ca. 400 CE to 800 CE]. The
Burgundians,
Goths, and
Vandals became linguistically assimilated by their respective neighbors by about the 7th century [CE], with only
Crimean Gothic lingering on until the 18th century." Another extinc Germanic language may also have belonged to the East Germanic family. "The 6th-century
Lombardic language . . . may be a variety originally either Northern or Eastern, before being assimilated by West Germanic as the
Lombards settled at the
Elbe."
Germanic Languages In General
All of the Germanic languages are descendants of the "
Proto-Germanic [language] (also known as Common Germanic), which was spoken in approximately the mid-1st millennium BC in
Iron Age northern Europe. . . . common innovations separating Germanic from
Proto-Indo-European suggest a common history of pre-Proto-Germanic speakers throughout the
Nordic Bronze Age." (Personally, I suspect that the pre-Proto-Germanic speakers arrived only around 1100 BCE in the late Nordic Bronze Age, rather than around 1700 BCE when the early Nordic Bronze Age begins.)
Proto-Germanic was a written language starting around the 2nd century CE when a runic script was used. Prior to about 750 BCE, the Germanic languages were spoken in an area roughly corresponding to modern Denmark and southern coastal Norway and Sweden. It only expanded into the modern boundaries of the Netherlands, German and other Germanic language speaking countries later on, reaching something fairly close to the current extent of Germanic languages in continental Europe by the 1st century CE.
Non-Germanic Languages In The Region and the Indo-European Languages Generally
The Non-Indo-European Languages Of Europe
All of the languages of Europe except Basque (a language isolate), Maltese (a derivative of Arabic) and the Uralic languages are part of the larger Indo-European language family.
The national language of the Scandinavian country of Finland is not a descendant of Old Norse and is not even Germanic or Indo-European. It is a member of the Uralic language family, the indigeneous language family of some of Northern Europe's last indigenous hunter-gatherers that also includes the Estonian and Hungarian languages. The Hungarian language is notable because this Uralic language is the result of language shift by a small Uralic language speaking elite that has left almost no genetic trace in the Hungarian population.
Baltic and Slavic Languages
The languages of Russia, Ukraine, Czeck Republican, Poland, Bulgaria, Slovakia, Slovenia, Macedonia, and Serbo-Croatian, in contrast, are Slavic languages. This Indo-European language family existed in the form of a single proto-language until about 500 CE (about the time that the Western Roman Empire collapsed), and then expanded from the general vicinity of the Balkans, replacing previous Indo-European and Uralic languages in the areas where Slavic languages are spoken now. They were differentiated into multiple distinct languages starting in the 7th century CE.
The languages of Lithuania and Latvia (and the now-extinct Old Prussian languages) are part of the Indo-European family of Baltic languages.
"All Slavic languages descend from
Proto-Slavic, their immediate
parent language, ultimately deriving from
Proto-Indo-European, the ancestor language of all
Indo-European languages, via a
Proto-Balto-Slavic stage. During the Proto-Balto-Slavic period a number of exclusive
isoglosses in phonology, morphology, lexis, and syntax developed, which makes Slavic and
Baltic the closest related of all the Indo-European branches. The secession of the Balto-Slavic dialect ancestral to Proto-Slavic is estimated on archaeological and glottochronological criteria to have occurred sometime in the period 1500–1000 BCE."
Romance and Celtic Languages
Many of the other major languages of Europe (e.g. French, Spanish, Portugese, Italian, Catalan, Occitan, and Romanian) are Romance language, i.e. languages descended from dialects of Latin that became distinct languages after the fall of the Roman Empire in the 5th century CE. The Romance languages are part of a larger Italic language family that also includes a number of extinct languages of the Italian pennisula. The Italic language family probably arrived on the Italian Pennisula from Central Europe sometime in the vicinity of Bronze Age collapse (i.e. about 1200 BCE).
The Celtic languages (e.g. Scottish Gaelic, Welsh, the Irish language, Bretton, Cornish and Manx) are more closely related to the Romance languages than any other living languages and the two language families combined are a genetic subfamily of Indo-European languages. Celtic languages were once spoken in territories that are now part of France, Spain, Portugal. The subdivisions of the Celtic languages started to emerge sometime between 1200 BCE and 800 BCE. The language of the late Bronze Age Urnfield culture of central Europe through about 1250 BCE was probably Proto-Celtic. The Iron Age
Hallstatt culture was definitely Celtic.
These languages are often lumped together as part of a larger
Italo-Celtic language family, perhaps with a proto-language in the Urnfield culture of its immediate predecessor.
Other Indo-European Language Families
The other living Indo-European language families are the Hellenic languages (i.e. a few Greek languages and many extinct languages), Armenian, Albanian, and Indo-Iranian. The Indo-European language family also includes the extinct Anatolian (e.g. Hittite), Tocharian and Paleo-Balkan language families. The Indo-Iranian languages are made up of:
* the Indo-Aryan languages of South Asia (and nowhere else except by recent migrants, by Balinese Hindu priests, and by the people colloquially described as gypsies) except in the Southern Indian areas where only Dravidian languages are spoken,
* the Iranian languages of Iran and neighboring areas the most widely spoken of which are Persian (75 million speakers), Pashto (50 million speakers), Kurdish (32 million speakers), Balochi (15 million speakers) and Lori (2.3 million speakers), and
* the Nuristani languages of about 130,000 mountain people of Eastern Afganistan and neighboring Pakistan.
Albanian is considered to have evolved from an extinct
Paleo-Balkan language, usually taken to be either
Illyrian or
Thracian. While Armenian is not a Hellenic language, it is more closely related to Greek than any other living language.
Indo-European Language Expansion
Indo-European languages arrived in Western Europe only in the Iron Age or perhaps a century or two earlier in some cases. Prior to around 2500 BCE, in my view, which is generally in line with the leading Kurgan hypothesis (the field has many competing hypothesizes about Indo-European linguistic origins), the Indo-European languages were probably absent from the Tarim Basin, from South Asia, from Anatolia, from Greece, and from Armenia. They were confined to the Balkans, Eastern Europe and Central Asia.
My personal and informed, but non-expert, opinion is that prior to Bronze Age collapse there was a copper age language expansion effected by the Bell Beaker civilization and its cultural descendants of languages that were part of the same language family as Basque, all of which (except Basque) were routed starting around the time of Bronze Age collapse by Indo-European languages. This copper age expansion probably caused the extinction of most of the pre-Copper Age languages of Western Europe and roughly corresponds with the geographic area where Y-DNA haplogroup R1b is common in modern European populations.