English Celticisms – the “Dummy Do” and “Obsessive Progressive”

Last time I discussed the influence of the Norse settlers on English during the period between Old English and Middle English and how it filtered down from the north to become standard English. About the same time, perhaps beginning a few centuries earlier, there was a Celtic influence that began in the south and filtered northward.

This Celtic influence is much more significant than the handful of words that I wrote about in the August 2011 post “Celtic Vestigia”, but is only now beginning to be generally recognized. 

Professor John McWhorter in Lecture 4 of Myths. Lies, and Half-Truths of Language Usage (Great Courses 2012) describes three Celtic influences that appear in modern English. Two relate to sentence structure, the third is a vestigial counting system.

Let’s deal with the numbers first. In northern England there is a special counting system called sheep-scoring numbers, presumably used originally to count sheep. This system today is found only in children’s games and nursery rhymes. It goes like this for 1-10: aina, peina, para, pedera, pump, ithy, mithy, owera, lavera, dig. The closest known number words to these are Welsh: un, dau, tri, pedwar, pump, chwech, saith, with, naw, deg. The system of teen numbering is also similar. Note these are not an exact match, but the 4, 5 & 10 are strikingly similar. So the origin of the sheep-scoring system was not, of course, Welsh but an older British Celtic language that somehow survived down the centuries only in this children’s number game.

As an aside, nursery rhymes are excellent “museums” of old words and phrases which are passed down intact while the language around them changes. “Pease Porridge Hot” is a good example of this. Pease was once the singular of the little green vegetable with peasen the plural. At some point the “n” dropped from the plural making both singular and plural words the same. Then, as the “s” ending of plurals became more common, someone assumed that pease was peas, the plural of pea, and this usage spread.

What McWhorter calls the “meaningless do”, Professor Anne Curzan in Lecture 9 of The Secret Life of Words (Great Courses, 2012) dubs the “dummy do”. Both refer to the English grammatical structure of using do in question and negative sentences. Instead of asking “Smoke you?” as would be normal in most other languages (including Old English), we ask “Do you smoke?” And we reply “I do not smoke” rather than “I smoke not”.

In Celtic languages, do is also used in positive statement sentences, but this is optional in English, used occasionally for emphasis. “Yes, I smoke” or “Yes, I DO smoke. Do you MIND?” More likely you would simply reply, “No, I don’t” or “Yes, I do” in answering a question like that, where the action verb in your reply is understood rather than repeated.

Back in Shakespeare’s day the do form of statements was more common. Hamlet, near the end of his famous soliloquy: “…thus conscience doth make cowards of us all”. And Duncan to Macbeth: “Glamis thou art, and Cawdor; and shalt be What thou art promis’d. Yet I do fear thy nature; It is too full o’ the milk of human kindness…”

Although this structure of using do sounds completely normal to us now, in fact it is very peculiar. The only other languages in the world that use do in this way are the Celtic languages. This use in English was first observed in Cornish English and gradually worked its way north. Cornish, spoken in Cornwall at the south-western tip of England, is one of the three Brittonic Celtic languages. Similarly to the core changes made by Norse speakers discussed last time, such changes to basic sentence structure were likely first made by Celtic (in this case Cornish) speakers who learned English as a second language. Bilingual speakers are more likely to use their native language sentence structure and overlay it with vocabulary from their new language. (I just made that up so would welcome comments from a linguist).

Similarly, the use of the progressive tense (the “ing” form of a verb indicating the action is ongoing) to show the present is another Celticism. This is what McWhorter calls the “obsessive progressive”. We ask “What are you doing?” instead of “What do you?” (in this case the verb do is the main verb, not a “dummy do”). And we answer “I am making dinner” not “I make dinner”. Again this is peculiar to English and the Celtic languages. It could be used in, for example, Spanish or German to emphasize that you are doing something right now, but that would be a rare usage (perhaps if your mother yelled at you “Get doing your homework!” you would reply “I AM doing my homework!”).

In Welsh to say “Mary is singing” it would be “Is Mary in singing”. Celtic languages have the word order VSO where the verb always comes first, so what looks to us like a question is in fact a statement. Notice the extra word “in” before the action word. English used to have this structure too but it was gradually lost with only a trace remaining in a few dialects. “I was on hunting” became “I was a-hunting” and finally “I was hunting”. See my post from October 2011 “A Hunting We Will Go” for more on this ancient grammatical structure. So this could be another Celticism.

Both the “dummy do” and the “obsessive progressive” show up suddenly in Middle English documents, with nary a trace in Old English. Why the sudden change? Two reasons – first there was 150-200 years starting after 1066 when French was the official language and hardly anything was written in English. Changes to the language during this time were not recorded anywhere. Celtic influence could have started as early as the 5^th century with the first Saxon-Celtic interaction. If so, why is there no record in Old English documents? Simply because writing at that time did not record how the common people spoke. Writing was used exclusively for church documents and high literature – it would have been unthinkable to record common speech.

So whenever you use the “dummy do” or “obsessive possessive” sentence structure, you are speaking a remnant of the language of the ancient British Celts. I think this is rather fascinating, don’t you?

The Simplification of English - Stage 2, The Viking Settlers

The first written texts of Middle English in the late 12th century show a remarkable simplification from Old English, just a century or two before. Verb and pronoun endings are reduced and simplified. Nouns lost their inflective endings (which indicate whether it was the subject or object of the verb) and relied more on prepositions and word order for meaning. Grammatical gender (nouns, adjectives and pronouns having different forms for male, female and neuter genders) was lost. And a few other simplifying changes that only a linguist could understand and appreciate. See this post by Linguistics Girl for a better explanation of these changes.

Linguists tell us that simplifications of this magnitude occur in a language only when it is learned as adults by a large enough group to influence the language spoken by the natives. It wasn't the Norman French, who were too few in number and mostly kept to themselves, who initiated this change. Rather it was the Norse from Denmark and Norway who had settled in eastern Britain in the late 8th and 9th centuries. This settlement (in contrast to previous Viking raids) was more or less peaceful. They ran businesses or farmed beside their English neighbors and often married English wives.

Because of their integration into English society, the Norse had to learn English. And adults, as you will know if you ever tried, find it difficult to learn another language fluently. Evidently they learned enough to get along and their simplified version of English gradually became the new standard across the country. Historical linguists tell us that these changes began in the northeast and gradually moved south over a period of several centuries.

A story related in Caxton's prologue to his translation of the Book of Eneydos in 1490 illustrates these changes in transition. A merchant ship was stranded in the Thames estuary by calm weather and a few of the men walked to a nearby house and asked the woman there for food. One asked for some "eggys", using the new (Norse) word. The woman didn't understand and said "Sorry I don't speak French", to which the merchant indignantly replied "Neither do I". Another merchant came to the rescue and explained that he was asking for "eyren", the old English word (singular "ey"), which the woman understood. You can read the story in Caxton's words here (page 2 starting line 25). He told the story to illustrate the predicament of English publishers in having to choose a dialect in which to print. It took over 200 years from the 14th to the 16th centuries for the word egg to become standard across England.

Norse also contributed many words to the English vocabulary. The most amazing are the pronouns they, them and their, replacing the Old English equivalents. Borrowing of pronouns is very rare and shows how significant the influence of Norse was on the core of the English language. Other core words include get, both, take, and want. Still other words are sky, skin, knife (and many other sk and kn words) and, as described above, egg. In many cases the Norse word was added, creating a synonym, rather than replacing the English word. Sometimes the adopted word took on a slightly different meaning, such as skirt (Norse) and shirt (English). Note that these words seem very English; they don't seem borrowed at all. Two reasons for this - Norse is another Germanic language so its words are more similar to English words than French or Latin words would be; and they were borrowed so long ago that they have become "Englishified" over time.

So English has two stages of people learning the language as adults and simplifying it in the process. First some people (possibly Semitic-speaking) in northern Europe learning Proto-Germanic (discussed in my previous post), then the Norse settlers in northeast England learning Old English. For them, all we native English speakers should be grateful, and those learning English as a second language even more so.

The Simplification of English – Stage 1, Proto-Germanic

While our children were in high school we hosted several European exchange students in our home for whom English was a second, third, or even fourth language. I was surprised when they told me that they considered English an easy language to learn, or at least less complicated than their own (French, Polish, German and Hungarian). What they meant by this was the lack of gender for most nouns and the simplified verb forms, not our beloved illogical English spelling.

John McWhorter, in “Myths, Lies, and Half-Truths of Language Usage” (The Great Courses), explains two quite different stages in the development of English where simplification occurred. The first occurred at the Proto-Germanic stage and therefore affected all the Germanic languages. The second occurred between Old and Middle English. In this post I’ll explore the first.

Compared to the other branches of Indo-European (e.g. Latin, Greek, Celtic, Persian, Sanskrit and Slavic), Proto-Germanic is different in several striking ways. One was the shift in consonants which is known as Grimm’s Law (discussed in an earlier post) where “p” becomes “f”, “k” becomes “h” etc. Another was a significant loss of complexity of verb forms. Where other IE language families have different verb forms for I, we (you & I), we (you, I & others), you (singular), you (plural) and they, PG had only one “we” and had the same verb form for “we”, “you all” and “they”. Verb tenses were also simplified to four (present, past, subjunctive and passive) from six in Indo-European. There were also some internal vowel changes in verb tenses introduced to Proto-Germanic (which I don’t understand so won’t attempt to explain here). Then there is vocabulary – many Germanic words were borrowed early on from the neighboring Indo-European languages of Roman and Celtic, but up to a third of Germanic roots are believed to be of non-Indo-European source.  What was going on here?

This type of language change (significant simplification) is not the normal pattern. In fact it is only seen when a language is learned by a group as adults. Who could these adults be?

First keep in mind that nearly all of the wide distribution of Indo-European languages is the result of imposition of the language on existing inhabitants by an introduced ruling class (either military or mercantile), rather than by mass migration of Indo-European speakers. See my post on the PIE Homeland. It had been proposed by Colin Renfrew (and others) that the Indo-European languages were introduced to Europe along with agriculture by farmers migrating west and north from Anatolia (modern Turkey). It made a nice neat theory but turned out not to be true – agriculture preceded IE languages by several thousand years in Europe. There were therefore, by the time the Indo-European language was advancing up the Danube into northwest Europe, farmers settled there with their own languages, whatever they may have been. It was these people who learned Indo-European as adults and adopted it as their own.

There is some linguistic evidence suggesting that these northwestern European people who learned Indo-European, and made it into Proto-Germanic, spoke a Semitic language (of which Arabic and Hebrew are members). Semitic languages are rich in the internal vowel changes of the kind seen in Proto-Germanic verbs; the fricative consonants “f”, “h”, and “th” introduced to Proto-Germanic as explained in Grimm’s Law are common in Semitic languages, and a few of the non-PIE Germanic roots seem to have a Semitic connection. This Semitic influence theory, promoted by Prof. Theo Vennemann, is quite controversial.

One theory to explain a Semitic influence on Germanic languages has the Phoenicians, a Semitic speaking sea-faring people who are known to have reached Portugal, to have sailed all the way around to north-western Europe – the Netherlands perhaps or even Denmark. This is highly speculative at best, as there is no archaeological evidence and so far to my knowledge no genetic evidence, to support their presence that far north. My theory (and you saw it here first) is that some influence of Semitic languages had accompanied agriculture as it made its way from the middle east to northwestern Europe. One of Bryan Sykes' maternal ancestors of modern Europeans described in his 2001 book " The Seven Daughters of Eve" is Jasmine who originated in the middle east and is associated with the spread of agriculture into Europe. Her genes spread into Europe and it seems quite likely that her language did as well, hundreds of years before the introduction ofGermanic Indo-European languages.

Whoever the first speakers of Proto-Germanic were, we owe them our gratitude for helping to simplify not only English but every other Germanic language spoken today.

The next stage of simplification for English, which occurred during the transition from Old English to Middle English, is credited to, of all people, the Norse Vikings who settled in northern England. But that’s a story for another week.