By Anatoly Liberman
It has been established long since that to know the origin of a word, one must know the properties of the object the word designates. This idea, sometimes neglected today (to the detriment of those who neglect it), dominated medieval etymologizing. For example, since God was universally understood to be good, people took it for granted that god and good are—must be—related (in fact, they are not). Conversely, for the statement that god, when read backwards, yields dog, witty heretics were burned. Apparently, certain principles were at stake. Modern historians and archeologists often turn to etymologists for assistance and occasionally get some advice from them, but in principle, it is word historians who need help. Consider the word thimble. There can be no doubt about the relatedness of thimble to thumb. Etymology, it appears, has the means to prove that thimbles were at one time meant for the thumb. However, this supposition should be modified. Etymology can only ask why a small cuplike guard used in sewing is called a thimble, though it is put on the third finger. Before our permissive epoch set in, female teachers in girls’ schools used the thimble (invariably worn on the index finger) to inflict painful blows on their charges’ heads. Therefore, it is a relief to discover that the older thimble, called “finger stall” or “finger bell,” indeed protected the thumb of those who pushed the needle through leather and other hard materials. Anyone interested in the history of thimbles will find a remarkable collection in the Fingerhut Museum (Kreglingen, Germany; Fingerhut, literally “finger cap”) and will understand how the thimble got its name. It follows, as predicted, that we need a look at the artifact to justify the etymology.
Everything is now clear, except for the spelling of thumb. No one has ever pronounced b in it, and it is absent from the word’s Old English form and from its cognates, such as Dutch duim and German Daumen. Since there is no museum of etymology or even a center for it and since dictionaries rarely dwell on such details, a post to explain the situation is needed. The irresponsible ancestors of modern English-speakers had the habit of inserting b between m and l. Bramble, grumble, mumble, nimble, scramble, shambles, number, humble, slumber, gambler, and quite a few others have the same parasitic (excrescent) b one hears in thimble (and in embers, b sprang up without l after b, probably in the group br). Unlike thimble, the words thumb, numb, and dumb have unetymological mb in spelling, though they do not end in l. Why do they? In principle, the story begins in late Middle English. At that time, consonantal groups were often simplified. Solemn and column are now pronounced without n. In the speech of some people kiln is homophonous with kill, g is always mute in diaphragm, phlegm, and so forth (the lost sounds have been preserved only when a syllable boundary separates them from their neighbors, as in solem-nity and phleg-matic). Because of the simplification, lamb was fleeced of its historical final b; today b is retained in spelling but not in pronunciation. The same happened to jamb, plumb, and tomb (borrowed words), along with womb (Old Engl. wamba “belly”; compare Wamba, the name of Cedric’s “fool” in Ivanhoe) and climb (Old Engl. climban). But timber already had parasitic b in Old English. French also sometimes has excrescent b in similar positions: so in chambre “room” and trembler “tremble” (from Latin camera and tremulare, evidently pronounced without the vowels given here in bold).
As long as the simplification of consonantal groups remained an active force, literate people felt uncertain when to write m, as opposed to mb, and began to add b to m gratuitously, a mistake (here, reverse spelling) called hypercorrection. This accounts for the modern forms limb (limber has excrescent b, regardless of whether it means “shaft,” “holes in timber,” or “pliable”; none of them is related to limb) and crumb. To be sure, it would have been more reasonable to abolish b where it was not pronounced, and this is what the related languages did (cf. German Lamm/Dutch lam and German dumm/Dutch dom for Engl. lamb and dumb), but English hates letting go of its orthographic relics. An 1819 dictionary still allowed its users to choose among jamb, jaumb, and jam. However, Samuel Johnson’s jamb won the day. It is hard to imagine that anyone even in the 15th century (which is indeed late Middle English) realized that numb and nimble contain the same root. Old Engl. niman “to take” had been superseded by Scandinavian take, and even if that had not happened, it would have been difficult to guess that numb originally meant “taken” (hence “insensible”), while nimble “quick and light in movement” went back to Middle Engl. nemel “able to take, grasp, seize.” Nimble was perhaps easier to identify than numb, for the verb nim had some currency until the 16th century. Later it degraded into a word of thieving cant (“steal”). Shakespeare knew it; otherwise, he would not have introduced Corporal Nym in The Merry Wives of Windsor. Thumb and thimble are probably still felt to belong together. If we ever reform English spelling, it may perhaps be reasonable to retain the spelling thumb, though I have great doubts about numb. But dumb is just dumb.
The change of ml to mbl is natural and common (the French parallel need not come as a surprise). Reverse spelling (crumb for crum, a word in which b has no justification) is also a trivial mistake. But everything is not so simple. From time to time we run into old words that have “organic,” etymological mb. Timber has been mentioned above. Another such word is comb. The sound b occurs after m in the Old English form and in all its Germanic and more distant cognates. However, not inconceivably, it was “parasitic” even 3000 years ago. The story of comb would not be worthy of mention, but for a closely related word no one today associates with it. The word is oakum. Old Engl. acumbe (with several recorded variants) has the same structure as Old High German achambi, and its etymology poses no problems: oakum, that is, loose hemp (earlier also tow, loose flax), fiber obtained by picking old rope, means “off-combings, what has been combed out”; a in acumbe was a prefix. We should be grateful to the phonetic change that treated acumbe so mercilessly. Otherwise, we would have ended up with oakumb. In dealing with the tomb of English spelling, one learns to appreciate every crumb of common sense.
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 firstname.lastname@example.org; he’ll do his best to avoid responding with “origin unknown.”