Yes, Your Wash-Up

The title is from The Posthumous Papers of the Pickwick Club (wash-up is the way one of the characters pronounces worship), but I owe the idea of this post to two questions. I decided not to wait for the next set of “gleanings,” because my summer schedule will prevent me from answering questions and responding to comments with the regularity one could wish for. Both questions concern Engl. r.

The first was about the status of r. Some people say that r is not a real consonant. Is it true? Yes, partly. In rate, late, mate, and Nate, the sounds r, l, m, and n, known in the phonetic nomenclature as resonants, are consonants like any other (compare Kate, pate, fate, date, etc.), but word finally they can form the crest of the syllable and thus display a feature characteristic of vowels. For example, each of the words—Peter, bottle, bottom, and button—pronounced as Pet’r, botl, botm, and butn, has two syllables, and the peak (crest) of the second syllable is r, l, m, and n. It follows that resonants sometimes behave as consonants and sometimes as vowels.

The second question requires a much more elaborate answer. Why do so many people pronounce wash as warsh? It will be easily seen that the title of today’s post was inspired by this question. In my recent discussion of wh-spelling, I referred to the weakening of Engl. h, s, f, and th, a process that has been going on for at least two millennia. Resonants are also prone to weakening. One can observe this change with a naked eye (with a naked ear?). British English is “r-less” (that is, r is not pronounced in far, for, fur, cart, horse, bird, and their likes), while in most varieties of American speech they are sounded. Obviously, the loss of r after vowels occurred in British English after the colonization of the New World by English-speakers, who preserved the traditional pronunciation of ar, or, ir, and so forth (colonial languages are always more conservative than the language of the metropolis). In similar fashion, l was lost in some positions, but that happened before the 17th century, so that here British and American English share a common cause: compare balk, talk, walk, chalk, balm, calm, alms, and salmon (in which mute l is still spelled) with bilk, whelk, and bulk (in which l is pronounced). The weakening was capricious: for instance, in Dutch a similar process took place, but the Dutch for holt “wood” (obsolete except as a last name) and salt is hout and zout: l is neither pronounced nor (providentially) spelled in them. In Scotland, golf used to rhyme with loaf (I don’t think anything has changed in the last fifty years).

After vowels, especially word finally, r disappeared not only in British English but also in some varieties of German. Occasionally other sounds, when weakened, find their last refuge in r. The name of the Greek letter that designated the sound of r is rho. Hence the term rhotacism “a change of any consonant to r.” Long ago, z was weakened to r. This is the most ancient case of Germanic rhotacism. The difference between was and were, raise and rear goes back to it; was ~ were and raise ~ rear are closely related pairwise. The weakening of t to d between vowels is a well-known phenomenon of American English. It accounts for the non-distinction of tutor and Tudor, futile and feudal, Plato and play dough, sweetish and Swedish, let alone writer and rider (cited in every work on American English phonetics). It is not for nothing that even well-educated people have been caught writing about deep-seeded prejudices, as though prejudices are basketball players. In at least one word this d (from t) was weakened further and turned into r. This is how pottage (as in a mess of pottage) became porridge. From an etymological point of view, pottage and porridge are doublets. In German and Frisian dialects, “rhotacized d” is more frequent.

This brings me to warsh and other cases of r appearing where it is not expected. Linking r is one of them. Even those who pronounce car and her without final r retain it before vowels, for example, in car and chauffeur and her and his. By analogy, they may say sofa(r) and bed, India(r) and China. Every sound change takes a long time and occurs through a generational shift. Thus, for example, some people may still retain r in car, cart, or, horse, burr, herd, and so forth, while others may already do without it. In such cases a feeling of uncertainty will arise in the speaking community the young and the old will be equally uncertain of which variant is correct. Both are acceptable, but one is “conservative,” whereas the other is “advanced.” As a rule, the “avant-garde” style wins out, though the “standard” may abolish the results of a change and return to the initial state.

It is easy to understand that when words like sort begin to lose their r, people, confused by the existence of two variants, will sometimes pronounce sought with r in the middle. Warsh may be due to this type of confusion. Readers of 19th-century British novels are familiar with the word arter, which reproduces the “vulgar” pronunciation of after. No doubt, Dickens and others heard r in arter; otherwise, there would have been no need to invent this spelling. Equally familiar is feller for fellow. The American verb holler may have r of the same origin as in feller. In light of these facts, warsh poses no problems: it is a trivial example of a pendulum swinging in the opposite direction. But strangely, pillow is often homophonous with pillar in British, Scottish, and Irish dialects in which r is not lost (both sound like pillar; such examples are numerous). Was r so weak that it could be inserted in unstressed syllables with impunity? Although wash has stress on it, the weakening of r even in those dialects in which it has survived must be responsible for the emergence of warsh.

The survival of individual forms centuries after (“arter”) a certain sound change seems to have run its course is natural. Someone who has grown up hearing warsh from the parents or classmates or both will not question the propriety of this pronunciation and will look on the written form wash as one of the many incomprehensible vagaries of English spelling. People who can say rite, need, and quire but spell Wright, knead, and choir cannot be expected to do anything reasonably. Later the necessity to please the teacher or adapt to the environment in which warsh is ridiculed may make such an individual drop the r, but those who learn a new norm as adults never feel comfortable with it, for deep inside they know that warsh is right and wash is wrong and have to control themselves every time they speak about soap.

Anatoly Liberman is the author of Word Origins…And How We Know Them as well as An Analytic Dictionary of English Etymology: An Introduction. His column on word origins, The Oxford Etymologist, appears here, each Wednesday. Send your etymology question to [email protected]; he’ll do his best to avoid responding with “origin unknown.”