Oxford University Press's
Academic Insights for the Thinking World


By Anatoly Liberman

Words, as I have noted more than once, live up to their sense. For instance, in searching for the origin of amaze, one encounters numerous truly amazing reefs. This is the story. Old English had the verb amasian “confuse, surprise.” Next to it, amarod (“confused”), the past participle of the unattested verb amarian existed. Both forms were rare, a circumstance that may perhaps bear out my final conclusion. In the entries on amaze, dictionaries of Modern English do not mention amarian, because, as I think, they have nothing to say about it and also because it has left no traces in the present day language. According to a well-known rule (Verner’s Law), s and r in Germanic alternated, depending on the place of stress in the protoform. If stress fell on the root, s was preserved, but, if it fell on the suffix or ending, s turned to z and later to r. Sometimes the “law” worked capriciously. For example, raise, a borrowing from Scandinavian, obviously had stress on the root (the modern pronunciation with z at the end is comparatively recent), while its native doublet rear must have been stressed on the suffix, no longer extant. Just why this happened remains unclear, but at the moment we are interested in the results, not in the mechanism of the change.

Similar alternations (s ~ r) sometimes occur in the forms of the same language, and I wonder whether the ancient verb amasian could sometimes have initial and sometimes “delayed” stress, obedient to the rules of intonation. Compare the situation in Modern English. Some people say ’fifteen rooms but Room fif’teen. Other than that, sentence stress fluctuates regularly: ’long a’go (both words are stressed), but ’ten ’years ago; we say ’Tennessee ’Williams but theState of Tenne’see.  I notice that the title of Thackeray’s novel has two stresses (‘Vanity ‘Fair), while the title of the magazine has one (‘Vanity Fair). Naturally, this is only a tendency. American speakers usually pronounce fifteen with initial stress in all situations, while some British speakers tend to distinguish between ‘fifteen and fif’teen. If my guess is right, there is no need to look for the etymology of Old Engl. amarod different from that of amasian. This hypothesis looks mildly attractive, but, as we will see, there is a stiff price to pay for it.

We now have to ask whether the second a in amasian was long or short. Old scholars preferred to look on it as long. Since long a in Old English had only one source, namely the diphthong ai (to which ei corresponded in German and Scandinavian), they cited as possible cognates Old High German meis “a basked carried on the back” and even German Meise “titmouse” (in Engl. titmouse, –mouse is the product of folk etymology: the titmouse is not a rodent, despite the ridiculous modern plural titmice), along with several ghost words. But amazement has nothing to do with birds or baskets, so that we probably have to reconstruct short a in amasian. And here we are lost: the spoor becomes cold. Skeat and Murray’s OED listed numerous Scandinavian nouns and verbs that begin with mas-. The OED online does the same. No one knows what to do with them.

An especially long list can be found in the old and excellent Norwegian etymological dictionary by Falk and Torp. Thus, we have Norw. mase “strive; bustle; beg; crush” (among other glosses, “lose consciousness” and “become delirious” turn up, but neither sense seems to be common). The dialectal and archaic noun mas means “whim; idle chatter.” Already here we begin to sense trouble: “strive” and “bustle” refer to a hectic effort, while “lose consciousness” presupposes listlessness. In Swedish, masa means “walk slowly, dawdle” and, in Norwegian, “warm (oneself).” At first sight, the two meanings have nothing in common. Quite a few related verbs refer to warmth, being drunk, and pleasant sensations, but “walk slowly” can be connected, if at all, only with “lose consciousness.”

It is the incompatibility of multiple meanings (striving and idleness, with warmth and intoxication thrown in for good measure) that baffles students of amaze and its look-alikes. In the nineteenth century, the greatest reward for an etymologist’s investigation was the ability to reconstruct the protoroot from which all the existing forms could be shown to have sprouted. In the first edition of his dictionary (under maze ~ amaze), Skeat set up the root MA “to think.” It is no wonder that he soon gave it up. Falk and Torp preferred the root MA “crush to dust.” Given enough ingenuity, one can derive any meaning from any other. “Crush” is better than “think,” for he who is “crushed” will occasionally lose control of his senses. Conversely, crushing may lead to bustling and striving. Today such exercises arouse subdued inspiration.

I would not have launched on the discussion of these intractable words if some time ago I had not come across a similar case, and also in Scandinavian. The root dras, perhaps known to some from the name of the mythological world tree Yggdrasill, has the same vowel variations as in mas and enters into the words meaning “rubbish; very big and heavy thing; pull with an effort; lazy; blunt; walk slowly, loaf; idle talk; indulge in debauchery.” (Drasill means “horse,” and it must have referred to a restive or uncontrollable horse.) Some of the senses are the same as in the mas– words. I would like to suggest that at least in the Scandinavian area rhyming slang words—dras and mas (and possibly more like them)—existed, none of which went back to a “respectable” Indo-European base. They seem to have had a broad and vague range of meanings, something like “impetuosity; willfulness; lack of predictability and direction” and could refer to about anything from working hard (and achieving nothing), loitering, lounging, and plodding aimlessly along to spending time with boon companions and lying in the sun. Old Engl. amasian would then end up as a borrowing from Scandinavian, because in English a comparable semantic network is absent.

Now the time has come to remember the stiff price mentioned at the beginning of this essay. Verner’s Law functioned very long ago, and if Old Engl. amasian and amarod are related by this law, they must have been current in English before the Germanic tribes settled in Britain. But I could not find similar words in Old Saxon, Low German, or Dutch. This weakens but does not derail my idea that amasian and amarod are related. Regardless of our conclusion, amasian, though lacking native siblings, and the numerous Scandinavian words appear to be congeners. Consequently, the isolation (within West Germanic) of at least amasian has to be accepted as an incontestable fact. But I would be happier if amarod did not exist! Since the participle means “confused,” my frustration deserves some sympathy. (“It is not customary to regret the existence of linguistic phenomena,” an unimaginative but solid professor wrote in a comment on my undergraduate paper.) However shaky my reasoning may be, I think that I am probably right in characterizing amaze as an example of old, perhaps even very old, slang, with analogs outside English.

A single note has to be added to what has been said above. When we see a word like aslant or askew, we assume that they are slant and skew, with the prefix a- being added to the root. Even askance and agog (about both of which I have written in this blog) must be a- plus some enigmatic skance and gog. But as far as we can judge, the English noun maze was abstracted from the verb amaze, rather than being its original stem. Maze denoted a place of utter confusion, which proves that amaze came into being with the sense “confuse, perplex” rather than “surprise.” The broad concept of surprise develops from various situations. For example, the inner form of surprise is overtake. Perhaps I will be able to return to this theme in the not too distant future.


Although maze is not the same as labyrinth, the distinction will be disregarded here. See the Minotaur trapped in the labyrinth. Stay amazed and happy.

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 on the OUPblog each Wednesday. Send your etymology question to him care of [email protected]; he’ll do his best to avoid responding with “origin unknown.” Subscribe to Anatoly Liberman’s weekly etymology posts via email or RSS.

Subscribe to the OUPblog via email or RSS.
Subscribe to only language articles on the OUPblog via email or RSS.
Image credit: The Minotaur. MAFFEI, P. A. “Gemmae Antiche,” 1709, Pt. IV, pl. 31. Public domain via Wikimedia Commons.

Recent Comments

  1. Graham Christian

    The creature in the labyrinth is a centaur, not the Minotaur. Citing a picture source–given the scholarly nature of the blog–would have been prudent, and told us more about the context. Centaurs often signified lust or ungoverned impulse or violence; likely this symbolizes man trapped in sin, but source citation would have helped.

  2. mollymooly

    @Graham Christian: See https://commons.wikimedia.org/wiki/File:Minotaurus.gif “The Minotaur is here represented (as not uncommonly in medieval and later times) as the reverse of the usual classical representation (a bull’s head on a man’s body). This is occasionally mistaken for a centaur; but the association with the Labyrinth makes it clear that it represents the Minotaur, and the artist has been careful to carve the hooves in bovine, not equine, form. The tuft on the belly is also typical of a bull, not a horse.”

  3. […] return to it, and, although the origin of astonish ~ astound ~ stun is less exciting than that of amaze, it is perhaps worthy of a brief […]

  4. […] November, Anatoly Lieberman, the etymology blogger for the Oxford University Press, chronicled the mysterious evolution of “amaze” from bygone Scandinavian tongues to Modern English with erudition and brio. He did not […]

  5. Mark S. Lawson

    In England, very strong cider called “scrumpy” used to be drunk from turned wooden bowls called ‘mazers’. Emptying one several times would definitely cause you to become

Comments are closed.