I am not a linguist. I do not know German nor French. The majority of English vocabulary is derived from Romance languages. Given these facts, I ask for a simple and convincing demonstration (using an example) that the "basic structure" of English is of Germanic, rather than Romantic origin.

  • 1
    English coalesced from the language dialects spoken by Germanic immigrants (Angles, Saxons, Jutes, Danes) who came to England from c. 450. – TheMathemagician Jun 10 '14 at 16:33
  • 1
    Your "Danes" were the Angles and Jutes. The fourth group was the Frisians (an island chain off what is now the Netherlands). I watched The Story of English (Jim McNeil) and it's amazing how close Frisian and (to a lesser extent) Danish is to English. – Phil Perry Jun 10 '14 at 17:09
  • 2
    If you look at examples of English prior to 1066CE, you'll find significantly less Romance influence. This is because it was during the Norman Conquest that a vast majority of borrowings and influence occurred during their rule. – Oso Jun 10 '14 at 18:27
  • 1
    Languages are usually classified by common ancestry. One could try to define synchronous language groups by select common features, as found in WALS, but I don’t think that leads anywhere. – Crissov Jun 10 '14 at 22:46
  • 3
    This is the first time I've ever seen an accepted answer with negative votes. – Miles Rout Oct 9 '14 at 0:20

There is a common assumption in all the answers so far, which I think is mistaken. It is that the question has a well-definied, categorical answer, that English is either Germanic or Romance, and cannot be some mix of the two. The assumption is clearly enunciated in some of the answers, but it is present in all, even in Janus very good answer, when he speaks of an early pre-viking "stage of English", as is a language has at any time T in the past a unique and well-defined ancestor.

We should remember that language have no parent like a bacteria has one parent and a human being has two. They are complicated structure with not so-well defined boundaries which evolves over time in part by borrowing a lot form other language. Most of the time, however, it is true that it is a very good approximation of the reality to suppose that a language has a unique ancestor at a given time of the past. For instance, it is relatively safe to say that the ancestor of Modern french at time 50BC is Latin, because even if there has been much vocabulary borrowing in French since 2064 years, it is marginal and most of the vocabulary and almost all of the grammars can be traced back from Latin.

But English has a very special history. At some point, French-speaking Normans invaded England and became the ruling class of the country. Progressively their French language mixed with old English of the former inhabitants, to give a new language which will in turn evolve into modern English. Now we need to turn to facts to see if this language comes mainly from Old English with some French influence, or from French with some Old English influence, or anything in between.

Most of the facts have been recalled by Janus, but let me reiterate them here and add new ones:

  • A majority of the vocabulary of Literary English comes from French while an important minority is of germanic origin.

  • A majority of the vocabulary of everyday spoken English comes from common Germanic but an important minority comes from French (for example, current words such as arm, table, chair, plate, car, flower, to push, to cry, etc.)

  • The grammar also shows the double influence: some traits, like the Saxon genitive "The man's car" are clearly Germanic, while some other, like the fact that most nouns take an s in plural comes from French. [This is to be seriously nuanced. See the comments.]

So I believe it is safe to say that English has both a Romance and Germanic origin, arguably slightly more Germanic in view of the importance of the Germanic words in its basic vocabulary, and of the form if Germanic origin in its grammar (as a more precise analysis would show). Remember that it is only a convenient approximation to say that a language is "the genetic continuation" of some language of the past, sometimes a very good approximation ("Polish comes from common Slavic"), sometimes just a decent approximation ("French comes from Latin", but 10% of its basic vocabulary is of Germanic origins, and some of its grammar too, like the "passé composé"), sometimes really an over-simplification ("English comes from common Germanic").

Added: For more information on a similar view, see this wikipedia page and the references given there.

| improve this answer | |
  • 9
    -1 As the other answers explain, the categorization of languages into families if historical and genetic, and only concerns how the language evolved rather its present state. – Anixx Jun 11 '14 at 20:20
  • 7
    "like the fact that most names take an s in plural comes from French." Whether French influenced the fact -s is now the only productive plural, I don't know, but the -s itself is definitely inherited from Germanic: etymonline.com/index.php?term=-s – dainichi Jun 12 '14 at 5:34
  • 5
    The answers aren't above anymore. Clearly the OP prefers their own interpretation of history and language evolution. In fact, the hardest things to change in a language are its grammar; words are stolen every day, but grammar changes like groweth to grows happen once a century or so, and so gradually nobody notices. So English is a rather irregular Germanic language, having gained 50% or so Latin vocabulary, and having lost almost all Germanic inflections, but retaining Germanic syntax and word formation. – jlawler May 3 '15 at 0:36
  • 1
    ... I am not asserting that middle English is a Creole of norman French and old English (I am not knowledgeable enough in germanic linguistic to venture in saying so, even though some serious linguists have done so), but my maine point, and I reiterate it, is that the most upvoted answers have failed to disprove (or even consider) this possibility, which may explain why the O.P. did not accept them. – Joël May 3 '15 at 16:18
  • 8
    I don't think the situation with English is that special. Japanese has a ton of words of Chinese origin, and currently there are a bunch of loanwords coming in from English as well, but this will never cause the language to be the same in origin as Chinese or English. Literary vocabulary is of no importance in the classifying of a language's origins. – brass tacks May 4 '15 at 7:24

Classification of languages is a historical thing, rather than a synchronic one. Just like genetic classification of humans—someone who marries into a new family and goes and lives with them is nonetheless still genetically related to the family they came from.

The majority of the total vocabulary in English may be borrowed rather than inherited, but the majority of the most common and basic vocabulary is inherited. This includes such things as numbers (all inherited), most pronouns (except they, which is Germanic, but not inherited), many basic non-administrative nouns (wood, name, stone, man, woman, ship, way, ox, hound, house, etc.), and many basic verbs (be, have, should, can, will, go, do, live, die, think, bear, etc.).

Much of this basic vocabulary is also among the most irregular in the language, which usually indicates it's been around much longer. Words borrowed from other languages tend to be force-fitted into the borrowing language's most regular morphology, while inherited words suffer no such restrictions.

If you go back to Old English from before the Vikings settled in England, you can clearly see a language (or several closely related languages, if you prefer) that has a high degree of resemblance to other West Germanic languages of the time, in almost every aspect. This (and the fact that this language can be reconstructed back to the common Proto-Germanic language that all Germanic languages go back to) is really the best direct indicator that English is genetically Germanic, rather than Romance. You won't find a stage of English where it is almost identical to any stage of any Romance language.

| improve this answer | |
  • 11
    Your wording suggests that it is irrelevant to classification how similar a language is to other languages today. I think it would help to add that vocabulary is quite a superficial trait of language (it's easy to borrow words from another language), and that when it comes to grammar or pronunciation (phonology, stress), English is much closer to other Germanic languages than to French. – reinierpost Jun 10 '14 at 10:29
  • 6
    @Morg. And yet, with the way English has evolved, there are quite a few features English now shares with French and Spanish, but not with, say, German (no cases, no grammatical umlaut, no distinction between reflexive and non-reflexive possessives, no verb-final word order, etc.). – Janus Bahs Jacquet Jun 10 '14 at 12:13
  • 1
    English was very simplified after the Norse invasion -- almost a Germanic pidgin. It lost gender cases, inflections, and accented letters (all of which Romance languages still have), among other things. The only connection with French and other Romance languages is a heavy overlay of French vocabulary. – Phil Perry Jun 10 '14 at 17:14
  • 4
    @Phil Accented letters is a matter of orthography; it's irrelevant to genetic affiliation (compare Hindi and Urdu: basically twin languages, but one written in Devanāgarī, the other in the Arabic script). Romance languages have all lost non-pronominal cases, too (except Romanian). – Janus Bahs Jacquet Jun 10 '14 at 17:35
  • 2
    @dainichi My German is not great, so you may be right; I thought paradigmatic umlaut was regular in a particular group of strong verbs in German, unlike in English where it's completely fossilised and appears only in a very small and random subset of words. And no, umlaut is not an orthographic concept: it is a phonological process that happens to be represented in a particular way in the (German) orthography. Ablaut is a different thing altogether. – Janus Bahs Jacquet Jun 12 '14 at 8:31

If you don't want to get into details of linguistics (which I take it you don't) the best way to see the family resemblance is to take a comparative look at English's closest linguistic relative found on mainland Europe: Frisian.

Some sample words in Frisian, English, Dutch, and German:

  • dei, day, dag, Tag
  • rein, rain, regen, Regen
  • wei, way, weg, Weg
  • neil, nail, nagel, Nagel

Frisian is of course indisputably a Germanic language, and just from the above its pretty clear both that these very basic words are all related, and that the Frisian variant looks far closer to the English than the other two.

As someone who has never learned other languages, it might be an easy mistake to think that vocabulary is all there is to a language. However, that would be wrong. There's far far more going on structurally in a language than simple word choice.

Delving into the murky waters of linguistics a bit more, we find that Germanic languages actually share a lot of pronunciation and structural features that are not found in Romance languages. Taking it further, West Germanic languages share features not found in North Germanic languages, and Anglo-Frisian languages share features not found in the other West Germanic languages. Based on all that, its fairly easy to classify English as Germanic, further as West Germanic, and further still as Anglo-Frisian.

| improve this answer | |
  • 2
    This context is really asking for the following Afrikaans sentence, which means exactly what you think it means: My hand is in warm water. Afrikaans is sort of a simplified Dutch with a little African influence and a different spelling system influenced by English. – user4938 Jul 29 '14 at 1:43
  • @HansAdler in Modern West Frisian this sentence would be "myn hân is yn waarm wetter", while the Dutch is "mijn hand is in warm water". De Afrikaans "my" choice of spelling leads to one more similarity to English, the rest is the same as Dutch. It's more like a pan-West-Germanic sentence.. – Henno Brandsma Jun 27 '18 at 22:10

Well, there are a lot of answers here, but I don't see the right one. We classify languages not by similarities of any sort other than shared sound changes in their histories. We inherit this geneological model from the 19th century neogrammarians, along with the assumption that sound changes apply without exception. Then what sound changes does English share with the Romance languages that other Germanic languages do not share? To my knowledge, none at all. There is no evidence to support classifying English as a Romance language.

Perhaps a case can be made that English shares some morphology and some phonology with Romance due to a large influx of loan words, including sets of morphologically related borrowed forms. But I don't know of any phonological rule of English traceable to Romance that could plausibly be classed as an exceptionless sound change.

Maybe the clarity of this issue is hard to see because so much doubt has been cast on the neogrammarian hypothesis of exceptionless sound changes. We could discuss that, I suppose, but I don't see it as an issue, here. That's the assumption that underlies the genetic classification under discussion.

| improve this answer | |
  • 2
    Not many comparative linguists today would claim that sound changes are exceptionless; rather, they are exceptionless unless there’s a good reason for exceptions to exist. It is also important to note that shared sound changes do not equal genetic affiliation, though the two concepts are not orthogonal either. Sound changes can be shared between non-related languages, and while we use clusters of sound changes to separate groups of related languages, that only really works going forwards, not backwards. Albanian, for example, shares few sound changes with other Indo-European languages→ – Janus Bahs Jacquet Nov 2 '15 at 22:55
  • 1
    @JanusBahsJacquet, there are some things there I don't agree with. Sound changes are phonological rules added to a phonological system, and so long as they are part of the system, they apply without exception in derived environments, though never in lexical environments. Phonological rules can be shared between non-related languages (being a Stampean, I'd say all of them always are), but I don't believe sound changes can. – Greg Lee Nov 2 '15 at 23:11
  • 1
    @JanusBahsJacquet, I was unclear. The "they" in my statement was intended to refer to phonological rules, not sound changes. A phonological rule is part of the phonological system of an individual. To be an exception to what I said, your example of s->h would have to apply for some people to some lexical /s/s but not others. (I don't really understand how an interjection or onomatopoetic word could fail to provide a lexical environment.) – Greg Lee Nov 2 '15 at 23:48
  • 1
    @JanusBahsJacquet, I made another mistake. I shouldn't have said that phonological rules generally fail to apply in lexical environments, but rather, that applies only to morphophonemic rules. So, for instance, it would constrain your example of s->h if the language had both /s/ and /h/ as phonemes. – Greg Lee Nov 2 '15 at 23:58
  • 1
    Ah, I see now. Yes, that is basically what frequently happens. Sound changes that generally occur without exception everywhere in a language are nonetheless often blocked in certain specific categories like onomatopoeia (e.g., Danish [b] → [b̥], but still voiced in baaa imitating sheep), interjections (Romance laughter having [h], otherwise lost), taboo, etc.; and occasionally for other reasons, like avoiding undesired ambiguity, too (e.g., a change like /l/ → /k/ in Greek would likely somehow skip the word pair καλός ‘good’ and κακός ‘bad’). – Janus Bahs Jacquet Nov 2 '15 at 23:58

For what is worth it is important to remember that all languages are, to one degree or another, blends of earlier languages. Some scholars have described modern English as a "creole" of old English and Norman French. Indeed some scholars have described the Romance languages of Western Europe (Italian, Spanish, French, etc.) as creoles of Latin and Gothic (Germanic). French, of course, got a second heavy Germanic blending as a result of the Franks. In the case of English, grammatically the language has been heavily influenced by French but, overall, its grammar is still somewhat more Germanic, though truthfully English grammar has morphed so much that a lot of it resembles neither French nor German. Certainly one can argue that English derives a lot more vocabulary from Latin than its Saxon roots but still the core of the language still resembles its Saxon roots more than its Latin influences (similarly one could point to the fact that Maltese has more Latin vocabulary than Arabic, but any linguist would tell you that, at its core, Maltese is Arabic).,

| improve this answer | |
  • I agree with this nuanced answer. +1. – Joël May 3 '15 at 16:19

No, Romance as a name is not preferable over Germanic.

To supplement the given answers let me add some meta-discussion about the concrete example.

I. classification is an ideal that does not represent the real world. It's a model of a theory, that is represented by English. A bunch of models represented by ideally individual languages make up a combined theory. That is bigger than just Proto-Germanic, and it's uncertain as much as the language is not actually ideal. This is especially true if our understanding of second languages rests on our understanding of the primary language. This counts all the more if we don't readily understand the language! And if we take a diachronic perspective, comparing Old Italian, Old English, Old French and Old German (with a representative dialect each?) then we are already two levels into the abstraction, so picking a one-dimensional naming scheme cannot be accurate. Henceforth, the one-dimensional thought needs to be interpreted on the first level. The zeroth level is just English. There is no exact threshold to distinguish levels like they were copper coins, but there is a pre-cised model of fuzzy matches to distinguish OE from ME, etc, so we run with it.

[BTW, can we make precised a word? Reasons for or against this are relevant to the question!]

II. The naive approach to a first level of comparison is naturally guided by the remaining mutual inteligability between certain subsets of each two Germanic languages, which rests inseperably on the grammar and lexicon--from the perspective of the English native speaker of course!

III. a) Vice versa, it's difficult to call English a romance Language. So far, it does not seem as if Vulgar Latin had any influence on early Old English. But that's precisely what it means to be Romance, if I'm not mistaken. Medievil, Classical Latin has not much to do with that. Arguing about the definition of Romance is besides the point--or too broad, here--because of the uncertainties due to the prevailing lack of attestation. Arguments to including English in that regard haven't been named in the question. It's also not strictly implied that the current naming scheme would preclude such arguments. Although it is expected. Nevertheless, such arguments should be welcome, if legal and justified.

b) If Romance evidence rested mainly on Italian, Spanish, French, and to a lesser extent the dialects inbetween, and further more the reconstruction including later Romance languages, then English seems a step removed, even if influence of Classical Latin is admitted, because it's peripheral, as far as I understand. Including English should strengthen the definition, in the mathematical sense, but it's not clear that it would--perhaps because this isn't maths; 1+1 doesn't always equal 2, even in maths.

c) If by Romance you mean a garbled mess, or vice versa a romanticized ideal, under Influence of Latin (sem)antics and (syn)tactics, then I'd agree, hell yeah; But it'd be a contradiction in terms. And what about Greek, why not Greek with all it's cultural influence. Yes, that's silly.

d) There are North European tongues that are hardly known or weakly attested. Their nature, and their influence on English, is speculative.

e) If you'd want to restrict yourself to Latinate French as the anchor, or at least include it in the picture, that's just shifting the goal post again. Influence of Frankish and Normanic (viz d) on French is non-negligable. Vice versa, given do-support in English, it should be deemed celtic--think you not?

IIII. Overall, much research needs to be done. The early history of the Brittish Isles, in this case the South of Great Brittain, as well as the continental coast is a puzzle in which language is a soft factor. Given west/north German evidence in South England, even Anglo-Saxon might be an understatement. Everyone who is serious enough to dive into details will hardly be bound by a prescribed label. If it can't account for all details, it's best to make a blanket term--a superficial statement to be made full by other definitions--that is however not meaningless as long as it encodes its own history. Germanicus is, at least as far as we can trace back, a Latin name, so it is not not [!] Romanic; rather Romanic in some sense--by virtue, if not by heart (any suggestions for an idiomatic aphorism?). More importantly, names are self fulfilling prophecies, so if we (read: I myself) call it Germanic, then because I want to keep entertaining the idea--Let's say it's not very influential a term at all just so I don't have to discuss socio-linguistic politics. But I have to admit bias. The blanket term has implications for the other Germanic languages as well. Is German with the hint of Italian, French, and Latin influence also a Romanic Language?

| improve this answer | |

Not the answer you're looking for? Browse other questions tagged or ask your own question.