Unpronounceable Words
As An Object of Etymology

By Anatoly Liberman

Unpronounceable and unprintable English words no longer
exist, and only the spellchecker turns red when smut defiles our screens for
academic purposes (dictionary makers and etymologists treat all words with
equanimity and, if needed, include and discuss them).  Now that everything
is discourse, taboo has been abolished.  Strangely, the more sensitive
people become and the more liable they are to take offence at an innocent joke,
the more often we hear the vilest mono- and disyllables from characters in
films and plays, commentators, and politicians. But the meaning of some words,
though not prohibited, has narrowed to such an extent that they now can be used
only in the most limited circumstances.  Older people still remember
gay “merry” (both gaily and gayety have survived into the
present, however), queer “odd, quaint,” intercourse “social
exchange,” and even ejaculation “interjection.”  Those meanings are
gone.

Other words are kept off premises not because they shock our
ears (nothing does) but because they happened to get into the way of
foul-mouthed neighbors. The family name of Immanuel Kant is one example.
To ensure the philosopher’s longevity, the English speaking world has adopted the
pronunciation can’t.  The result is both more seemly than the
“proper” pronunciation and still sounds German.  One feels a bit
embarrassed when calling someone a nincompoop, in some cases a most appropriate
word with an unsavory (and particularly apt) termination.

Not too long ago, the adjective niggardly caused a
storm.  All of a sudden etymology found itself in the limelight, and those
who ply this trade realized the blessings of obscurity.  I talked myself
hoarse explaining the origin of niggardly, trying to save its
reputation.  But can it be that some of our readers are still unsure of
how that adjective came about?  Since it is better not to take chances, I
will repeat the thrice-told story below.

In the Old Germanic languages, many words began with gn-,
kn- and hn-, and the first letter was not mute.  These
consonantal groups have disappeared in English, though the conservative
spelling of knock, gnaw, and the like reminds us of the reality of gn-
and kn- in the past (hn- is avoided in Modern
English).  No plausible hypothesis accounts for the simplification of
initial kn- ~ gn-.  English speakers can pronounce acknowledge and
ignominy (the latter with stress on the first or the second syllable),
while students repeat German words like knapp “short” without the
slightest difficulty.  In canoeing, one often hears knuing,
and yet the name of King Knut has turned into Canute.  Knish seems
to hold its own well, though the pronunciation with a short vowel between k
and n has also been recorded.  As always, one detects a trend and
puts up with exceptions.  Knob, knock, knot, and gnaw
have retained their medieval spelling.  Other similar words are spelled
phonetically today, so that without consulting an etymological dictionary one
cannot guess that nap (in take a nap) traces its ancestry
to Old English hnappian or that the Dutch cognate of nibble is knibbelen
“haggle.”

Language historians are lucky when they have the support of
forms like hnappian and knibbelen, for those make an earlier
stage of nap and nipple partly clear.  But many words are
isolated.  In British English dialects, knurd occur whose meaning
is reminds one of nerd.  Is it possible that several centuries ago nerd
began with kn-, or is the spelling with kn- fanciful?  Such
questions beset us all the time.  The only known cognates of
nod (which surfaced in English texts in the 14^th century) are
German notteln “move about, shake” (regional) and its etymon (source) notten.
German has partly lost g/k before n (for example, the German for gnaw
is nagen).  The original form of notten and, consequently,
of nod may have begun with gn-; however, there is no
certainty.  Similarly, nudge, whose earliest attestation in English
goes back to 1675, probably developed from some Old English form like hnycgelan
or cnyccan, but here, too, we are not sure.

Many words with initial gn- ~ kn- ~ hn- seem to have
designated light, sometimes repetitive movements (consider the meaning of gnaw,
nibble, nod, knock, and nudge).  It is unclear whether we are
dealing with sound symbolic or sound imitative formations.  The line
between the two types is often hard to draw, and their age is almost impossible
to ascertain.  The end of the 17^th century seems to be too
advanced a date a coinage like nudge.  It only turned up in books
in 1675: the length of its “preliterary” existence in dialects is beyond
recovery.  Local and slang words flourish in informal speech and gain
wider acceptance only when someone promotes them.  Nudge was both
dialectal and slang.  Niggard, first recorded in Chaucer (who had
his ear attuned to popular speech), is akin to niggle “to do anything in
a trifling or ineffective way” (from gniggle ~ kniggle?) and is, most
likely, a borrowing from Scandinavian.  There is a good chance that at one
time it sounded like gniggard or kniggard; hn- does not count: it
is only a weakened form of them both (for instance, knife has a
counterpart in Icelandic, and there it begins with hn-).   If
I am right, niggard was a name for a penny pincher, a miser who would amass
wealth in a “niggling,” painstaking way.  Being called a niggard, that is,
stingy, is not a compliment, but not an ethnic slur either.  Its use is a
matter of speech aesthetics rather than semantics.

Anatoly Liberman is the author of Word Origins…And How We Know Them. 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.”