A New Linguistic Process?

Download PDF of this post

Here’s a book with a novel thesis: the English language has been, to coin a term, Vikingized. A creolized language is a grossly simple pidgin language that has been turned into a syntactically-complex mother tongue. A Vikingized language is the opposite: a syntactically-complex natural language that has been turned into something simpler. Creolization reflects the natural ability of children to communicate richly. Vikingization reflects the natural difficulty adults have learning the subtleties of a foreign language. I find myself intrigued by the idea because I can immediately think of two other languages—Swahili and Afrikaans—that are candidates for the same historical process. If investigation supported that thought, and other Vikingized languages can be identified, John McWhorter’s book (coming out in paperback tomorrow, Nov. 3)—Our Magnificent Bastard Tongue: The Untold History of English—will have proved to be a real breakthrough in linguistic theory.

The trouble is, McWhorter does not prove his thesis. It would take a detailed look at the history of English to prove it and this book is nothing close to being a history of English. It is a collection of five essays, a couple of which have most interesting historical theses and a couple of others that are as familiar as sporting clichés. It leads me to suspect that McWhorter does not quite understand what he has got here and, like a newspaper reporter who is new to his beat, has buried his lead. He thinks his story disproves the old Whorfian hypothesis that language shapes our thoughts, and perhaps it does, but that is old news. The really important thing is that it offers a new and potentially important category of language change. Unfortunately, I have to write about the book that exists, not the one I’d like to read. So lets’s look at the two interesting essays. 

Oh, and there is an introductory chapter that makes an interesting point as well. McWhorter says that the history of English is much more than the history of words. It is also the history of grammar and he wants to focus on that part. That sounds promising.

The opening chapter then says that two of English grammar’s most prominent features —the use of a “meaningless” do in many questions, and the addition of –ing to form the right-now tense—come from the Celtic inhabitants of England who preceded the Anglo-Saxon invaders and brought Old English with them. 

The –ing suffix is especially interesting.  McWhorter points out that grammar books tend to say that the present tense conjugation of to write is I write, you write, he/she/it writes, we write, you write, they write. And yet if you ask me what I’m doing, I do not say, “I write my blog post,” but, “I am writing my blog post.” I write, says McWhorter, is not really the present tense, but is used for things that occur regularly. If somebody asks what I do for a living, I can reply, “I write,” but that is not happening this instant. 

Regrettably, the story is not as simple as McWhorter says. At one point he writes, “…English speakers have to say He is feeling better where almost all the other Germanic languages would say He feels better.” [p. xxi] That brought me up short. I think I remember my mother saying, “He feels better,” and she was a native speaker of English, albeit with a southern drawl. I checked on Google and found 1,470,000 listings of “he feels better” and only 74,300 listings of “he is feeling better” [quotation marks included in the search]. So sometimes the form is optional and sometimes required. There are also times when it is forbidden. Indian-English speakers commonly say things like, “He is wanting a ride,” which, outside the Asian sub-continent, is just a mistake. But why and how did these rules sort themselves out?

That would make a great story. Too bad McWhorter doesn’t tell it. Instead we learn that the Celtic languages (Cornish, and Welsh) insisted on using the progressive form of the verb for present and passed the form over to English. We should learn that fact, and I’m glad to have finally found it out, but instead of then tracing in detail how this form worked and changed over the centuries we get scores (scores!) of pages answering objections to the idea of Celtic influence. I was persuaded long before McWhorter stop trying to convince me. 

The other interesting account comes in chapter three, where McWhorter describes the Vikingization of English. Briefly, for a few hundred years during the darkest of the Dark Ages, the British Isles were subject to repeated Viking colonization. Norsemen arrived, took (Old) English-speaking wives and spoke broken English around the house. Both Old English and Old Norse were Germanic languages that declined nouns, had gender, and elaborate conjugation markings. But the English and Norse markings were different, and the Vikings were unsure of the English system. The result of a couple of centuries of invasion was the stripping of gender, declension, and verb markings to a bare minimum. This process of simplification made English much easier to learn and transformed it from a complex, Germanic language into a Vikingized tongue that foreigners can pick up readily. 

It is a neat story, and I’m open to it, but some things have to be made clear to me. First, the Anglo-Saxon invaders seem to have been exactly the sort of people the Vikings were—male invaders who took up with the local women. So why did the Viking descendants speak English (of a very bad sort) while the Anglo-Saxon descendants did not speak a Celtic language? McWhorter does address the lack of Celtic words (chance), but does not go into the larger issue of why Celtic receded to Cornish and Welsh corners while the populations stayed put. The problem is not insurmountable, and might be resolved in a full blown history that looked at how the Celts and Anglo-Saxons interacted versus how the Vikings and English interacted. 

McWhorter might well reply that it would be great to do studies like that, but they are impossible. We don’t know anything about the subject. It is all buried in the mystery of the miserably documented centuries immediately following the fall of Rome. In particular, the spoken languages of those centuries are undocumented. Most of the people were illiterate, and writing was the work of trained scholars. The Old English they wrote was a kind of classical language, using genders, declensions, and verbal conjugations that may have disappeared from the actual speech. We go with the feeble information we do have. It’s a fair point, but then we really have to try to get every piece of information we can find, and it is not sufficient to tell us that English first came under the influence of Celtic co-existence and then Viking stumblings. That information does not even get us to William the Conqueror. 

I want to know particularly how English recovered from its Vikingization. If the Viking families spoke some kind of near-pidgin, then some kind of creolization-like recovery should have occurred. In fact, we do know what happened. Declensions disappeared, but word order became much more fixed. Gender was lost, but a new he/she/it system came in that distinguishes between the quick and the dead, and the sexes of the living. Conjugation became simpler, but a complex system of auxiliary verbs and prepositions took its place. I’ve always thought of those things as a kind of co-evolution, but if they were a recovery from a “battered grammar” I’d like to know how the recovery worked.

McWhorter could answer reasonably enough that he does not know how it worked. But he could have at least surveyed the questions and possible answers. To do that, however, he would have had to know he had piqued some curiosity. Instead he thinks he’s telling us that “there is nothing unique about English’s ‘openness’ to words from other languages … [and] there is no logical conception of ‘proper’ grammar as distinct from ‘bad’ grammar” [171]. The first point is obvious once made and the second has been made so many times over so many decades that I’m heartily bored by it. But Vikingization—that’s a new idea for me, and one I will keep in mind.

Comments

[JanetK]

I may be out to lunch here but... I was taught at some point in my life that the Danish Vikings as opposed to the Norwegian came from the same area as the Angles but some hundreds of years later. They settled around the Humber and in East Anglia where the Angles had settled previously. Therefore the roots of words in the local English and the Danish Viking were the same. Instead of learning one another's grammar they both came to use the root words unadorned. If this is true, it would be an example of deducing two languages with elaborate grammars to a single language with a simplified grammar. I have a little distrust of such neat little explanations but the general idea seems feasible. It doesn't account of the half of England that escaped the Danelaw or the place where the Norwegian Vikings settled or the places that were settled by the Saxons or the Jutes or Friesians. etc. etc. They are all similar Germanic languages and the process of creating a single language would have been long and complex but a root word pidgin may have been the starting point many times and places.
----------------------------
BLOGGERS: I too wondered why the practice spread beyond the Danelaw area. McWhorter notes that it was slower to spread beyond but does not really give us a process for the spreading. I have the impression that the Scandanavians in general spoke mutually intelligible languages,

[JanetK]

Sorry - typo - deduced should be reduced.

[Uzza]

Tech-nit-ally, I write and I am writing are both present tense, but differ in Aspect, being Perfective and Imperfective respectively. That is doubtless what McWhorter means by by “not really present tense”. Habitual actions take the imperfective form in English.

Obviously other languages don't say the English sentence “He feels better”, but I read page xxi as saying German uses the perfective form where English uses the imperfective (aka progressive) form. (*these terms are not standardized)

Incidentally, “He be wanting...” in African American Vernacular English is correct form for imperfective aspect.
-----------------------------------
BLOGGER: What's the story on "He is wanting a ride." Is there something in Hindi/Urdu that supports the practice?

[uzza]

Re Hindi, Yeah. English has a rule that you shouldn't use the progressive form with stative verbs. Most North Indian languages don't have such a rule, and speakers of Indian English have just ignored the restriction. That lets them use “I am wanting ...” where Englishmen have to use “I want...”.

The AAVE case is different. A common process in Creole formation is: grab a common lexical item, grammaticalize it into an auxiliary verb, and use it to express a distinction in the Tense-Aspect system that wasn't made previously. That's how 'invariant be' got into AAVE, to express habitual aspect: they can say “He be wanting ...” where Standard American has to say “He always wants ...”; and both still say “He wants” for the non-habitual meaning.

[David Fried]

An interesting thesis, but not new. I remember it being advanced in the PBS series "The Story of English" with Jim Lehrer quite a few years ago. For whatever it's worth, I think the loss of case endings is so unremarkable in European languages that this explanation is superfluous. I also think this is just a special case of "creolization," which is, after all, McWhorter's speciality.

One has to take into account that the Viking settlement was almost immediately followed by the Norman Conquest, which quickly ended the use of the English vernacular for official and literary purposes. I'm no expert, but I understand that the formal literary Anglo-Saxon still in use at the Conquest was already a learned tongue, 2 or more centuries behind the vernacular. It thus preserved many forms no longer in use.

The new standard that began to take shape 200-300 years later was based on different dialects than Anglo-Saxon and came straight from the vernacular speech of the time. So, in effect, 4 centuries or more passed between the formation of the two standards. That case should have been lost in the interim is not surprising and requires no special explanation.
--------------------------------
BLOGGER: Most of this post is a good summary of the McWhorter thesis, but (using McWhorter's dates), the Normal conquest was about 200 years after the Viking period's end. You can decide whether that is a short or long time.

[David Fried]

I am unfamiliar with the Normal Conquest, but it sounds like a bad idea.
---------------------------------------
BLOGGER: I think I'll leave that one uncorrected. Quod scripsi scripsit.

[JanetK]

I think David is right in his time line, and in the hidden development of separate local English dialects for a couple of hundred years after the Norman conquest. I am interested in why David says that losing case endings is a normal thing in European languages. Are there other European languages that have lost case and gender endings almost entirely or even substantually? If so were they creolized too at the time of the loss?

[David Fried]

Janet,

Thanks for the support. After all, William the Conqueror came along less than a generation after all England was ruled by the Danish King Knut--the "Canute" who ordered the tide to recede!

As for the loss of case--the only Romance language to retain a simplied case system is Romanian, and that seems to be less a survival than the influence of other Balkan languages. The Latin case system quickly disappeared from Italian, Spanish, French and Catalan. The loss of case in those languages is plainly an internal development, resulting from the disappearance of final "m", the sign of the accusative, a process that had already begun in preClassical times; the loss of vowel length, etc.

Classical Arabic had a case system, too. But its loss may be probably attributable to the "creolization" of Arabic when it became the language of tens of millions of people as the Muslims conquered much of the known world. Lots of those people spoke Aramaic and Syriac (pretty much the same thing), Semitic languages related to Arabic that had lost their case endings more than 1,000 years before.
---------------
BLOGGER: Isn't "creolization" being used here in the sense I've been using for vikingized? There are two processes: (1) building up from a pidgin to a creole language; we can call that creolization; (2) tearing down, from a complex language like classical Arabic to a simpler one that is easier to learn; we can call that vikingization.

And once you start thinking of vikingization, it's hard to stop. Latin looks like a good candidate since its empire was famously invaded by a very large number of barbarous tribes. The Normans from France had been Vikings a bit earlier, so I wonder if they spoke a vikingized French. And what's the story with Norman Italian? Was it vikingized as well?

[David Fried]

Right--I agree that the process of rapid simplification and change that hit Classical Arabic was "vikingization" as you're using it. But the point is that Arabic and Aramaic largely coincided in their triliteral roots and general structure--the morphological complexity of Classical Arabic was partly lost in the shuffle. But when the Vikings gave up Norse for Norman French, nothing happened to French, as far as I know. Nor would you expect "vikingization"; French and Norse are just too different. And the "barbarian invasions," if they can even be said to have occurred (see a terrific recent book called "The Ruin of the Roman Empire") had remarkably little effect on the language of the Roman West except in England, where Latin was killed off. Again, the barbarians simply gave up their languages for proto-Romance, just bringing in a handful of lexical items like "guerra," (obviously cognate to "war.") The common features of the Romance languages must have been present in the Vulgar Latin of the late Empire.

[JanetK]

Maybe it goes like this. If the two languages are different then a pidgin is formed; if the two languages are similar then a vikingized language is formed. In effect the speakers are going to retain as much as they can of their former languages in some sort of spectrum between drastic pidgin-ing and mild viking-ing. I suspect that we would also have to take into account the power inequalities between the two peoples.

[L]

The reason McWhorter didn't push "Vikingisation" as his discovery is that it's neither his idea nor a particularly new idea. A good discussion of it with specific reference to English can be found in one of the later chapters of Thomason & Kaufman 1988: Language contact, creolization, and genetic linguistics. The term "creolisation" is quite commonly used to mean what you call Vikingisation; this is arguably a more approriate meaning for it than Bickerton's, since there is no direct evidence that most creoles ever went through a stage of being pidgins.
-------------------------
BLOGGER: I used 'Vikingization' to distinguish it from Bickerton's 'creolization.'