Recent Focused Series »

Indo-European Origins
Northern California
The Caucasus
Imaginary Geography
Home » Europe, Indo-European Origins, Linguistic Geography

Is It English or Engelsk?—part 1

Submitted by on January 7, 2013 – 12:55 am 21 Comments |  

In recent months, GeoCurrents has been concerned with an article about the origins of Indo-European languages, published in Science and publicized in the New York Times and other media outlets. It has become amply clear from that project that when it comes to matters of language and linguistics, the popular media’s “scientific” reporting typically sensationalizes studies that make outlandish (and often unsupported) claims, while ignoring other work on the same topic. In the next few posts we will look at several articles on specific Indo-European languages that have recently received much attention in the press and the blogosphere, starting with a revisionist history and classification of the English language.

ScienceNordicIn the last decade or so, renewed attention has been paid to the history of English in academic circles, leading to the publication of several new textbooks by principal academic presses. Cambridge University Press came out with A History of the English Language, edited by Richard Hogg and David Denison, as well as a much more accessible The English Language: A Historical Introduction by Charles Barber, Joan C. Beal and Philip A. Shaw. Oxford University Press countered with The Oxford History of English, edited by Lynda Mugglestone. Scholarly monographs and trade books in the field have been published in large numbers as well. Yet none of them has received as much public attention as a yet-unpublished manuscript by Jan Terje Faarlund of the University of Oslo and Joseph Emonds, visiting professor from Palacký University in the Czech Republic, who claim that English is a Scandinavian language. The story was originally broken by ScienceDaily, which justly characterized the claims made in the paper as “sensational”. Other media outlets in the English-speaking world, such as The Economist and Business Insider (which reposted the piece from The Economist under a different headline), were more tempered, formulating the headlines as questions or using the modal may. But Scandinavian websites as well as several English-language blogs came out with resolute headlines. ScienceNordic went so far as to illustrate their piece with a picture of prince William and Kate Middleton wearing traditional Scandinavian sweaters, and Aftenposten, a leading Norwegian daily, simply affirmed that “English is a Scandinavian language”. The title of the k2p blog post is more detailed but equally firm: “Modern English derives from Scandinavian rather than from Old English”.


Unfortunately, I have not been able to get hold of this semi-mythical paper, as it has not yet been published. According to media reports and a detailed interview with Jan Terje Faarlund, the two scholars deny the received wisdom that Modern English descends primarily from Old English, or Anglo-Saxon, which was a West Germanic language, most closely related to Old Frisian.* Instead, they propose to classify (Modern) English as belonging to the Scandinavian (or North Germanic) group, together with Norwegian, Danish, Swedish, Icelandic, and Faroese. Faarlund is cited in ScienceDaily as saying:

“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. […] We believe it is because Old English quite simply died out while Scandinavian survived, albeit strongly influenced of course by Old English.” [this latter statement is contradicted by another statement quoted from Faarlund, as we shall see below]

As a result, the authors contend, Modern English is quite different from Old English, and also from modern West Germanic languages, such as German, Dutch, Frisian, Afrikaans, and Yiddish. As is well known, many English words do not match their German (or Dutch) counterparts: compare the English die and German sterben, or the English ill and the German übel. These English words derive from Old Norse deyjaand illr, which replaced the Old English words steorfan and yfel. (The Old English words did not die completely, however, but survivedas starve and evil.) It is well-known that the Norse-speaking Vikings gave the English many words which they took and still use, whether that seems both odd and wrong to some fellows (the boldfaced words in this sentence are all Norse contributions). In a sentence like The guests cut the rotten cake with a knife, only the articles the and a are not traceable to Scandinavian sources.** Among Norse loanwords in English are basic kinship terms (sister, husband), body parts (leg, neck, skin), other common nouns (dirt, sky, window), adjectives (flat, loose, ugly), and verbs (drag, get, smile). Northern English dialects, spoken in areas that once constituted the Viking-dominated Danelaw, contain even more words from Old Norse, such as fell ‘hill, mountain’ (compare with Norwegian fjell) and kenning ‘knowledge’ (compare with Swedish kännedom ‘understanding, cognizance’). In northern English cities like Leeds and York, toponyms with -gate like Briggate and Kirkgate translate as ‘Bridge Street’ and ‘Church Street’, because in Scandinavian gate means ‘street’ (in contrast, in London places such as Aldgate and Newgate actually refer to former gates in the city wall).

But Faarlund and Emonds place emphasis not so much on words that have been borrowed from Old Norse but on English grammatical structures that do not exist in German or Dutch. Their claim is that “wherever English differs syntactically from the other Western Germanic languages—German, Dutch, Frisian—it has the same structure as the Scandinavian languages”. Four examples have been cited to bear on this issue:

1) Word order: Both English and Scandinavian place the object (underlined) after the verb (boldfaced), whereas German and Dutch put the verb at the end.

English: I have read the book.

Norwegian: Jeg har lest boken.

German: Ich habe das Buch gelesen.

2) Preposition stranding: In both English and Scandinavian the preposition (boldfaced) can appear at the end of the sentence, with its related noun or pronoun (underlined) placed in the beginning.*** This structure is much more restricted in German or Dutch.

English: This we have talked about.

Norwegian: Dette har visnakket om.


3) Split Infinitive: In both English and Scandinavian we can insert a word such as an adverb (boldfaced) between the infinitive marker to and the verb (underlined). This structure is also highly restricted in German and Dutch.

English: I promise to never do it again.

Norwegian: Jeg lover å ikke gjøre det igjen.





4) Phrasal (or “Group”) Genitive: The possessive marker in both English (‘s) and Scandinavian (s) can appear after a whole phrase like the Queen of England, not after the main word (i.e. head) of that phrase (here, Queen).

English: the Queen of Englands hat / Mom and Dad’s only child

Norwegian: Dronningen av Englands hatt / mor og fars eneste barn

Roman_Roads_in_Britannia.svg But Faarlund and Emonds go further than to simply note the similarity between the English and Scandinavian constructions: they claim that the English grammatical morphemes and structures were adopted from Scandinavian and survived to this day, while “Old English quite simply died out”. They seek further support for their theory in geography, noting 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”. Indeed, Matthew Townend provides the following description of the linguistic situation in medieval Britain in his chapter in The Oxford History of English:

“Spoken Norse appears to have been both geographically widespread and surprisingly long-lived, no doubt because it formed the first language of a substantial immigrant community. Settled Norse speakers were to be found in England from the 870s onwards, following the Viking wars of the time of King Alfred (who reigned over Wessex 871-99) and the establishment of the so-called Danelaw; that is, the area to the north and east of the old Roman road known as Watling Street […] Norse continued to be spoken in the north of England certainly into the eleventh century, and quite possibly into the twelfth in some places.”

While the influence of Norse-speaking Vikings on the English language is undeniable, there is no solid evidence for the claim that English is a Scandinavian language. The first thing to note is that by comparing Modern English with Modern German and Norwegian (which perhaps was done to get the point across to the general public), Faarlund has committed the cardinal sin of historical linguistics: examining more recent forms of language rather than the oldest available forms. A much more convincing comparison would be between Old English, Old High German, and Old Norse. Such historical investigations show the close affinity of all West Germanic languages, including both Old English and Old High German, as reflected in a number of phonological innovations, such as the development of numerous diphthongs in positions where North Germanic languages have a pure vowel and a consonant. For example, the Old Norse hoggva (and Modern Swedish hugga) correspond to the Old English verb hēawan ‘to cut, hew’ (the diphthong is also evident in the modern German hauen); similarly, Old English brēowan ‘to brew’ corresponds to Old Swedish bryggja, Modern Swedish brygga (and the diphthong is retained in German brauen). A number of lexical items also point in the same direction. According to Barber, Beal and Shaw,

“one lexical form found only in West Germanic is the word sheep (Dutch schaap, German Schaf, Old Frisian skēp), which has no known cognate elsewhere. […] the Old Norse word was fār (Old Swedish) or fǽr (Old Icelandic): the Faores are the ‘Sheep Islands’ (Old Icelandic Fǽreyjar)” [p. 90]


But could Faarlund and Emonds be correct in identifying the four syntactic structures listed above as Scandinavian loans? Curiously, R.L.G., the author of the language blog in The Economist, notes that two of the four patterns mentioned above—namely, split infinitives and preposition stranding—“are controversial in some usage circles”. While most people use these patterns in natural speech and English writers have done so for centuries, some “traditional but half-informed pedants claim that you can’t split an infinitive or end a sentence with a preposition”. Just how pervasive such “incorrect” structures can be is seen from the fact that a leading prescriptivist of his time Robert Lowth, who penned A Short Introduction to English Grammar in 1762, committed the very sin of preposition stranding as he was criticizing it:

“The Preposition is often separated from the Relative which it governs and joined the verb at the end of the Sentence … as, ‘Horace is an author, whom I am much delighted with.’ … This is an Idiom which our language is strongly inclined to; it prevails in common conversations, and suits very well with the familiar style if writing; but the placing of the Preposition before the Relative is more graceful, as well as more perspicuous; and agrees much better with the solemn and elevated style.”

The long-established explanation is that prescriptivists thought that English should imitate Latin (or Ancient Greek), where such structures were impossible, but maybe, R.L.G. ruminates, they “rub some people the wrong way because they are bad Anglo-Saxon, not bad Latin”. As tempting as this explanation would be, it does not work. First, not all presumed Scandinavian grammatical imports are problematic for prescriptivists: while the Queen of England’s hat may sound strange to some people, the Queen’s of England hat is much worse. Moreover, there is nothing remotely questionable about the VO order in Modern English; in fact, if you say I have the book read, you are likely to get funny stares.

But more importantly, Old English already had the seeds of all four of these constructions. For instance, preposition stranding—which was first ascribed to Scandinavian influence by Logeman in 1906—was attested in Old English in a restricted set of contexts such as with personal pronouns (in the following example, ‘me’):

Þa wendon hi me heora bæc to
then turned they me their back to
‘Then they turned their backs on me.’



So the change from Old English to Middle (and hence, Modern) English was not in the introduction of a completely novel structure, but in a change in the range of contexts where it is possible. The Old English option of separating a personal pronoun from its preposition, as in the above example, died out soon after 1200. Around the same time, preposition stranding became possible with passives (as in ‘dealt so cruelly with’), and later extended to relative clauses (e.g. ‘the book which we have talked about’) and questions (e.g., ‘Who did you talk to?’).

In the following GeoCurrents post, we will consider the other syntactic structures presumed to be of Nordic origin, as well as the more general issue of whether languages borrow grammar and under what conditions it is most likely to happen.




*Recall that the Gray-Atkinson model (see the image on the left from Bouckaert et al. 2012) incorrectly classifies Frisian as most closely related to Dutch, not to English.




**The word with existed in Old English in the form of wið meaning ‘against, opposite, toward’; this older meaning is preserved in compounds such as withhold, withdraw, withstand. The influence of the Old Norse vidh is responsible for the meaning shift in Middle English to denote association, combination, and union. In this meaning, with replaced the Old English mid ‘with’, which survives only as a prefix, as in midwife, literally ‘woman who is ‘with’’ (the mother at birth). The original sense of wife ‘woman’ (regardless of marital status) is also preserved in the expression old wives’ tale.

***When it comes to preposition stranding, one has to be careful at distinguishing true prepositions (which can be stranded) from particles (whose syntax is quite different, as evident from the humorous hypercorrection This is nonsense up with which I will not put, attributed to Winston Churchill, or alternatively to Bernard Shaw).



Previous Post
Next Post

Subscribe For Updates

It would be a pleasure to have you back on GeoCurrents in the future. You can sign up for email updates or follow our RSS Feed, Facebook, or Twitter for notifications of each new post:

Commenting Guidelines: GeoCurrents is a forum for the respectful exchange of ideas, and loaded political commentary can detract from that. We ask that you as a reader keep this in mind when sharing your thoughts in the comments below.

  • Peter Rosa

    If it’s all right to ask something only peripherally related to the post …
    I’ve heard that German is a particularly good language for engaging in debates or arguments, because the placement of the verb at the end of the sentence means that it’s difficult to interrupt the speaker in mid-sentence (until you hear the entire sentence you won’t know what he or she is trying to say). Does this actually make sense?

    • Asya Pereltsvaig

      Yes, of course it’s alright, Peter!

      I understand precisely what you mean and I’ve heard claims like this before. But it only makes sense if one assumes somehow that the verb carries more information than any other word in the sentence: just as a German hearer has to wait till the end to find out what the verb is, the English hearer has to wait till the end to find out what the object is.

      There’s another problem with this idea: there’s many languages around the world that have the verb at the end of the sentence. In fact, Subject-Object-Verb order is the most common basic word order in the world! This list includes Japanese, Korean, Hindi, Yakut, Ket, Khoekhoe… Check out the map here:

      Are these languages also good for arguments and debates? I’ve never heard the advocates of German-as-best-debate-language claim that. As a matter of fact, the abovementioned languages should be judged as even better than German, because they always have the verb at the end, unlike German, which, as you can see in that WALS map, is not a uniformly SOV language, but a Verb-Second language, meaning that in German the verb appears in the second position, not at the end, in main clauses. So in German, ‘Hans read the book’ will be literally ‘Hans read the book’, not ‘Hans the book read’, which would be the literal translation of the corresponding sentence in Japanese, Hindi, Yakut, etc.

      Finally, this reminded me of a joke: two grad students in philosophy went to a guest lecture by some very famous German philosopher, but despite his reputation, the lecture turns out to be very boring… they listen for a while trying to make sense of it, and finally one of them says, “Let’s just leave, it’s so boring” and the other one replies: “Let’s give him the benefit of the doubt and wait till he gets to the verb”.

  • O.T. Ford

    I know you’re not claiming to have seen the original article, but the main point of it, you note in the second and third paragraphs, is that Modern English is not a descendant of Old English. I share your skepticism about this assertion, and would need a very persuasive article to give it much consideration. But when you introduce your counterargument (in the paragraph beginning “While the influence of Norse-speaking Vikings”), you seem to be begging the question (which, for the benefit of other readers, I am using to mean “assuming the thing you are trying to prove”). You say that comparing Modern English to Modern High German and Norwegian is misleading, and we should instead be comparing their ancestors, “Old English, Old High German, and Old Norse”. But then isn’t the claim of the mysterious article that the West Germanic dialect of Old English is -not- the ancestor of Modern English, or any other modern dialect?

    • Asya Pereltsvaig

      You make an excellent point, O.T.! However, my general point, which still stands, is that one should compare varieties that are as old as possible, in order to control for the possibility of later independent developments or borrowings. So the best comparison would still include Old High German and Old Norse, as well as Old English (and possibly Middle English, although there was some 200 years between the arrival of the Vikings and the beginning of the Middle English period). In fact, one might be able to get away with considering only Old English (and not Middle English), by looking at pre-Norse and post-Norse varieties. My main argument against Faarlund’s alleged claim is not that Old English was more similar to Old High German (and even more so to Old Frisian) than to Old Norse, but rather that the specific syntactic structures presumed to be Norse imports are not that, but are rather natural developments from (pre-Viking) Old English. This is what I am trying to show in this post and in the two posts to follow, to appear tomorrow and the day after, if all goes well.

      Come to think of it, it is interesting that the pre- and post-Viking varieties of English did not get distinct labels (the same way that pre- and post-Norman varieties did)—that alone shows that the differences are not that significant.

      • O.T. Ford

        Indeed. And doesn’t the lineage of the speech community trace to the Anglo-Saxons, not the Norse? It can’t be enough to show that the Anglo-Saxons borrowed words or even structures. There would need to be a point, a discontinuity, where they simply adopted wholesale the speech of the Norse, or we would still be compelled to say that English evolved out of the pre-Norse Anglo-Saxon speech. Surely there are enough sources to show the gradual change, including gradual adoption of Norse items, rather than the punctuated expansion of Old Norse in the Anglo-Saxon population.

        • Asya Pereltsvaig

          You are absolutely right that in most cases the abrupt language shift would be clearly identifiable. The confounding factor in the case of English is, however, that Old English and Old Norse were really similar to each other, so the transition, had it happened, would be much more difficult to identify. But I will talk about this in the two follow up posts, so stay tuned!

  • James T. Wilson

    Is it really so surprising that only the surprising is news? Were I to hear that astronomers had found that the moon was slightly more massive than previously thought, I would yawn. Were I to hear that they had found that it would be colliding with the earth in a century or two, I would take notice.

    I have always wondered about the split infinitive in English. As I recall, the infinitive is a single word in Old English (e.g. heawan above), so it could hardly be split. Was there some use of a preposition with modal verbs or something, or did this form some later, when the infinitive came to be constructed from two words?

    • Asya Pereltsvaig

      In the same vein, were you to hear that biologists (not astronomers!) discovered that the Earth is flat, would you care to read on or would you just say that such nonsense isn’t worth your time? So it’s not just surprising or even sensational that’s the problem but the nonsensical, opinionated, unsupported that bothers me.

      • James T. Wilson

        Oh, I would certainly read on, but for much the same reason you would, I’m sure.

        • Asya Pereltsvaig

          Yes, sometimes one just needs a reason to get mad at someone ;)

          • James T. Wilson

            I suppose, then, that my reaction is more anthropological than yours. I am always interested in what people believe and why they believe it. My own worldview is so different from any other I have ever encountered that I can usually not only agree with the famous phrase of Terence, but also with it’s opposite: Alienus sum, nihil humani similis mihi puto.

          • Asya Pereltsvaig

            Oh, after I get mad at them, I can start analyzing what people believe and why… :)

      • James T. Wilson

        Do you know anything about that split infinitive business? I have always heard this from English teachers who say this was some rule taken over from Latin, but Latin wouldn’t need such a rule, since the infinitive is one word, nor, I imagine, would Old English.

        • Asya Pereltsvaig

          Ah sorry, I missed that part of your earlier comment.

          As far as I know, there were two infinitive-like constructions in Old English: the actual infinitive was a one-word verb, consisting of a stem ending in -n or -an, and there was also the so-called gerund consisting of “to” + a verbal noun in the dative case, which ended in -anne or -enne (e.g. tō cumenne = ‘coming, to come’).In the Middle English period the two forms coalesced, creating what we have in Modern English: “to” (from the gerund) + bare stem (from the infinitive). There are no known examples of split gerund in Old English that I know of, but that might be a preservation issue. Once the two structures coalesced, split infinitives start to appear.

          The prescriptivist rule that infinitives must not be split appeared much later (prescriptivism in general doesn’t appear as such until 17th century). At that time many of the “rules” about how English should be were indeed based on the (somewhat idealized) view of how Latin was, so since in Latin infinitives were not (and could not!) be split, the same was assumed to apply to English, erroneously of course, since it’s structure is completely different.

          Hope that clarifies things!

          • James T. Wilson

            It does, although if it is not found in Old English, I do not know why this rule would be any more foreign to the structure of English than any of the other syntactic oddities that sprang up in Middle English but would sound quite foreign to modern ears. In all modesty, English is a language I have been speaking with some fluency for a number of years, and the split infinitive sounds as odd to me as an adjective coming between the “zu” and the verb in German would.

          • Asya Pereltsvaig

            Adverb, you mean? I think in some cases infinitives can be split even in German. I might be wrong on that though…

  • Lane Greene

    Having written the Johnson blog post you refer to, I just want to make it clear that I too concluded that a “Scandinavian theory of English” is neither necessary nor sufficient.

  • Herman Callens

    At one point in your text you remark that “the Gray-Atkinson model (…) incorrectly classifies Frisian as most closely related to Dutch, not to English”.

    This, indeed, appears to be the general view, but it is based on the Anglo-Frisian situation of a long time ago. Today, Frisian is much closer to Dutch than to English. Wikipedia will tell you that: “(…) that Frisian has now far more in common with Dutch and the adjacent Low German dialects, bringing it into the West Germanic dialect continuum, whereas Anglic has stronger North Germanic and non-Germanic influences than the languages on the mainland” (see And if that won’t do (after all, it’s just Wikipedia …), one can always find Frisian/Dutch/English texts to compare, e.g. here: Clearly, Dutch and Frisian vocabulary are more alike, and word order is almost completely identical. As a Dutch speaker myself, I have no difficulty whatsoever to figure out Frisian (any more – and sometimes less – than a Dutch or Flemish dialect), which I have never studied, but I would never have understood half as much of English without having studied it first. Frankly, to me Frisian looks more like a Dutch dialect than like a language of its own, but I wouldn’t dare to say so to the Frisians …

    So, maybe it is not such a big surprise that the Gray-Atkinson model comes up with their Frisian classification, even though, in an older stage, the language situation was different (slightly, for that matter, as the languages at that time hadn’t been drifting apart for very long). As for the present situation, their classification of Frisian is in fact correct.

    • Asya Pereltsvaig

      Exactly. But the phylogenetic tree (family tree) is meant to represent the historical relatedness between languages, not current similarity. Which just confirms that they got that wrong!

      • Herman Callens

        I had a quick look at some word lists that were used for the Gray-Atkinson model, based on (meanings of) “basic vocabulary items”, as you know, “thought to be relatively universal and resistant to borrowing” (as the authors put it, and I won’t go into the pros and cons of their ‘choices’ here). For some reason that seems to put Dutch and Frisian closer together. Maybe because in the range of “basic vocabulary items” Dutch and Frisian happen to be more alike, or maybe because English, Frisian and Dutch were much more ‘intertwined’ (the “continuum” I talked about in my previous post), making clear cut boundaries too artificial to justify the “rightness/wrongness” of whichever model.

        But the situation may be worse than you think. In some cases the lists do contain different words for a given meaning, whereas native speakers immediately see that (striking) similarities have been left out. Take ‘bad’, for example. There are two Frisian words in the list: ‘erch’ and ‘lilk’. The Dutch word given is ‘slecht’ and the Old English one ‘yfel’. That looks all very different, doesn’t it? However, Frisian ‘erch’ is the same as Dutch ‘erg’ (‘bad’ in the sense of ‘serious’, ‘grave’) and Frisian ‘lilk’ is Dutch ‘lelijk’ (‘bad’ in the sense of ‘ugly’). Old English ‘yfel’ is today’s ‘evil’, but also Dutch ‘euvel’.

        Unfortunately, Old Frisian and Old Dutch forms are not listed, but let’s look at them, too. Frisian ‘erch’ and Dutch ‘erg’ are Old Frisian ‘erg’ and Old English ‘earg’. To my knowledge, modern English doesn’t have a similar word. Dutch ‘lelijk’ is derived from ‘leed’ (Old English ‘lað’, related to ‘loath’) + the suffix ‘lijk’ (Middle Dutch: ‘leetlijc’), Old Frisian ‘lethlik’. Again, no such word in English now. Finally, Old English ‘yfel’ is Old Frisian ‘evel’ and Old Dutch ‘uuil’ (to be read as ‘uvil’). All three languages still have it today: ‘evil’, ‘evel/euvel’, ‘euvel’.

        Okay, that is just one example, but I could give quite a few more. All in all the lists do seem to suggest that Dutch and Frisian are a bit more similar than English and Frisian, although, let’s not forget, the differences are relatively minor. I think we should be careful then and not jump to conclusions about who is ‘right’ or ‘wrong’ on the basis of such scant evidence. I am certainly not ruling out the possibility that the model is (in this case, at least) more right than wrong, unlikely as it may seem. But whatever it is, we are talking about a detail here which, at best, is only marginally relevant to the qualities (if any, I’ll add …) of the model, and maybe even completely irrelevant. Clearly, the Gray-Atkinson model
        is anything but a final word. When it comes up with unexpected or ‘wrong’ answers, it might be good to wonder why: maybe the model/method is no good, maybe our ‘answers’ are no good or maybe there is a flaw somewhere that can be explained when looked into. At least in the case of Frisian, our ‘answer’ may not be as sound as we would
        wish. Which is not to say that the model is sound. Further research may bring all answers together.

        • Asya Pereltsvaig

          Thanks for pointing out some of the problems in lexical comparisons of English, Frisian and Dutch. They only underscore the points I’ve made in the post and elsewhere: that the vocabulary may not be the best comparanda for historical linguistics…