LINGUIST List 2.429

Fri 23 Aug 1991

Disc: Why "wh-"

Editor for this issue: <>

Directory

1. , Why wh-?
2. Geoffrey Russom, Re: Queries: click, why wh-
3. Bill Poser, wh words
4. "Charles A. Bigelow", Re: Queries: click, why wh-

Mark Sanderson <sanderso%dcs.glasgow.ac.ukRICEVM1.RICE.EDU> asks:
 >Does anyone know why in English words questioning words
 >(I'm afraid I don't know the correct term) like
 >
 > Who, What, Why, When
 >
 >all begin with "wh" ?
 >Is there a good reason ? Do other languages have this quirk ?
Yes, Mark, there is a reason. As to whether it is a "good reason", opinions
differ. And, yes, other languages have this quirk, though not all in the
same way.
The English interrogative and relative pronouns and adverbs are technically
called "wh-words", even when they aren't spelled (like how) or pronounced
(like who) that way, and even when not dealing with English, at least by
some people. They come historically from the Indo-European forms, all of
which began with a /kw-/ consonant, pronounced moreor less like "QU-" is
in ordinary English. You can see this in Latin, which preserved the
sound - the Latin forms all begin with "qu-"
 qui - who quis, quid - what quo - how, when, where
...et cetera. If you put a "wh" wherever you see a "qu" here it begins
to look awfully familiar. How did that happen?
Well, English is a Germanic language, and a particular kind of sound change
happened to the Indo-European sounds as they mutated into the Germanic
languages. It's called "Grimm's Law" because it was discovered by the
brothers Grimm, who did other things besides collect fairy tales. Grimm's
Law says that all the "stop" consonants (which includes /kw/) change to
other consonants in various regular ways. In particular, what happens to
an Indo-European /kw/ is that it changes to /hw/, which was probably
pronounced like a German "ch" followed by an English "w". In English
this sound got softer and softer over the years, so that many English
speakers today don't distinguish between the first sound in (say)
"where" and "wear", even though they're spelled differently.
I said that opinions differed about whether Grimm's Law was a "good" reason
in the sense you asked. It's certainly true, which is a benefit; and it
does explain where the English forms came from. But it doesn't explain
two things:
 1) Why the Indo-European "wh-words" (more properly, "kw-words")
 all seemed to have this peculiarity; there are explanations for
 this, but Grimm's Law isn't one, since it merely says where
 they came from, not how they got that way to start; and
 2) Why Grimm's Law happened the way it did. Grimm's Law is descriptive,
 not explanatory. In fact, nobody's got the foggiest idea *why*
 sound changes occur the way they do, though there are some theories
 as to why they should occur in general. This is approximately akin
 to a theory of evolution without a concept of natural selection.
 So when you hear linguists talking about languages and words being
 "genetically related", remember it's just a metaphor, and not a very
 good one at that.
Good question. Thanks for asking.
 John Lawler jlawlerum.cc.umich.edu
 Linguistics Program jlawlerumichum.bitnet
 University of Michigan - Ann Arbor

The "wh" words of English correspond to "kw" words in Latin (quis,
quid, qui, quare, etc.). It's an Indo-European quirk. I don't know
what IE constituent corresponds to the "kw" part, but I think it may
be a separate word originally. A parallel case I do know about involves
glitter, gleam, glisten, glow, and other words having to do with light.
The "gl" in these words is zero grade of IE GHEL, the root of "gold" and
"yellow."

The reason that English "what", "who", etc. begin with "wh" is that
in Proto-Indo-European the corresponding words began with "kw", of
which "wh" is the regular English reflex. Compare Latin "quod", "qui",
etc. This kind of regularity is not unknown in other languages. In
Japanese, for example, we have:
koko 	here		doko	where
kore	this		dore	which (pronoun)
kono	this		dono	which (adjective)
kare	him/her		dare	who
konna	this sort of	donna	what sort of
kotira	this of two	dotira	which of two
ko:	in this way	do:	in what way
The k(o)- terms (here by me) have corresponding sets in s(o)-
(there by you) and a- (there away from us both), with some
exceptions and irregularities I won't go into here.
However, some wh-words don't fit: "what" is "nani", "when" is "itsu".
It looks like this is enough of a paradigm to induce analogy.
Historically, the wh-words have initial /d/'s because they
began with /i/, e.g. Old Japanese idoko/iduko "where". (Native
Japanese morphemes almost never begin with voiced obstruents.)
Howver, the old form of dare "who" is /tare/. The voicing of
/tare/ to /dare/ is attributed to analogy with the other wh-words.

The historical linguists can tell the tale more competently and
completely, but in case they don't reply, here is a non-expert
account:
Most of the English interrogative pronouns in "wh-" are, in some dialects
pronounced like /hw-/. And this is the clue to their origin. (Though the
"wh-" in "who" is an exception pronounced /h-/, and in some dialects,
mine included, "wh-" is reduced to "w-", so that "when" is homophonous
with "wen" and "where" with "wear". A great shame, I'm sure.)
Anyway, the /hw-/ (or sometimes /w-/ or /h-/ is all that remains in
English of the late great Proto-Indo-European interrogative labialized
velar /kw-/. It is seen and heard in Latin "quo" ('whither'), "quis"
('who'), etc. and is still resoundingly popular in modern Italian and
Spanish. It also appears in Hittite, which, however, isn't spoken much
these days. By the series of sound changes known as "Grimm's Law", which
maps P.I.E. consonants to their reflexes in Germanic languages, the /kw-/
becomes /hw-/ in modern English (as /k-/ -> /h-/, /p-/ -> /f-/, and so
on. (I show them here as initials, but, subject to some more context
sensitive rules, the sound shift also applies to the consonants in
other positions.) The elucidation of such diachronic phonetic mappings
was the pride and glory of 19th century linguistics.
Standard textbooks of historical linguistics can provide full and
frank discussions of Grimm's Law and other excitements, including
laryngeals (sounds which, in an astonishing case of predictive linguistics,
were hypothesized by Saussure as existing in P.I.E. even though none
survived in any known language, until their discovery in Hittite,
the cuneiform texts of which were not unearthed until several decades
after Saussure publication.)