[ This essay appeared online in The Vocabula Review in August, 2005. I've revised it a bit here. --JK]
e like to believe we say what we mean, we serious people, with
an exact grasp of denotation. But the truth may be that we choose our phrases
first for sound, letting meaning sift in rather casually.
Infants understand this. During the first year, while nonverbal
cries, smiles, frowns, gestures, and eye contact suffice for all substantive
communications, their real language development begins as a love affair with
pure sound. The spontaneous crib talk that charms grownups in all cultures has
a vital purpose: it allows babies to explore their own sound-making and
sound-detecting abilities while they tune in to the phonological system of what
will soon be their native language. Parents pace and fret, eager for the first
dispatch from the great beyond that is baby’s consciousness: “C’mon, Brittney,
say ‘Daddy’; say ‘car’; say ‘transmogrify.’”
But Brittney knows instinctively that meaning will wait.
What really matters is phonology, the essential foundation on which everything
else will be built. So she goes on patiently babbling, guided only by her
pleasure in rehearsing the basic sounds and sound combinations she hears around
her. Such practice is one of the best educational investments she will ever
make, helping her attain the perfect native pronunciation that forever eludes those
who learn a second language as adults.
(Think of Fareed Zakaria, Arnold Schwarzenegger, Guillermo del Toro, Yoda.)
When the words finally do start to emerge from the stream of
babble, they seem as doubtful as faces in the clouds. We might not count them as
words at all, except that culture seems to force the issue, taking the sounds
baby is most likely to make anyway
and assigning the least unlikely meanings to them. Hence in language after
language we get Dada or something
like it for the father, Mama or
something like it for the mother, and Baba
or something like it for the little one herself. Sound comes first. Once the
syllables are there, reliably
produced and distinguished, we can begin to hang meanings on them — but not
before.
It is, of course, always a question of multiple meanings.
Words are just like that: endlessly recyclable, always ready to surprise us by
saying something they have never quite said before. Humpty Dumpty is wrong when
he tells Alice, “When I use a word, it means just what I choose it to mean,”
and offers the example of making “glory” mean “there’s a nice knock-down
argument for you.” But his error is only one of degree. What a word can mean is roughly stipulated by the
culture, but what it does mean in a
particular instance depends on all the particularities of context, on a
creative and personal collaboration
between speaker and listener. All this is true from the very beginning, by the
nature of the case, and so before long it occurs to Brittney that the dog is a type
of Mama, and she points and says so.
We laugh, but know to at least a first approximation what she means: “This,
too, is a large, pleasant creature whose purpose is to amuse and take care of
me.”
Let’s take a moment to emphasize this initially arbitrary, wide-open quality of words. A cardinal feature of human
language, everyone notices, is the way we combine and recombine sounds
meaningless in themselves to make infinite combinations, laying the groundwork
for what becomes a huge lexicon with amazing conceptual reach and subtlety. The
word “fish” consists of three sounds, none of which has any meaning in itself,
but that is exactly the key. If the lush sh
by itself had meant “fleeting silvery good-to-eat lives-in-water thing,” it
could never be used in other combinations; but this way, the phoneme can play
an indispensable role in composing thousands of different words.
So far so familiar; but less often noticed is that, even
when you have fish, you are still a
long way from really determinate meaning. Are we including crayfish and
silverfish? How about whales and dolphins? Do we mean a particular one, a school,
or a class? What of the
fact that it can be not a creature at all, but an action, and shouldn’t fishing
logically be called defishing? What
about metaphorical extensions like big
fish, cold fish, fishing in her purse, and so on? What if I turn the word
into an adjective, with a reasonable expectation that fishy must mean “edible and nourishing”? All such matters await resolution
in particular contexts. Fish as it sits
there in the dictionary, atop a dozen glosses, is mainly a blob of potential, a
vague sketch.
But a better way of putting it is that the word by itself is
a sort of atom whose true potential is not realized until it enters various
combinations. In effect, the original trick of “coherent arbitrariness” at the
phonemic level gets repeated at the lexical level, with a similarly handsome
payoff. Just as sh means nothing in
itself, but comes to mean a great deal when it is the sound that distinguishes fish from fin, so fish itself
becomes meaningful mainly as it is put into harness with other words: the fish is already a huge step toward
determinacy, and the fish you ate last night
seems perfectly clear for all ordinary purposes, though a philosopher could
still argue various uncertainties. Like the phoneme, what the word has
initially is not meaning, so much, but just identity, a place in the system. As
long as everyone hears it as something different from fin, fib, fix, flush, and
so on, we’re good to go: meaning can come in later, rather gradually.
So on top of phonological indeterminacy, we have lexical
indeterminacy; and then much the same thing happens again at the phrasal level.
As foreign language students always learn, a huge portion of the phrases most
indispensable to any language count as “idioms”: combinations whose meanings
are not really the sum of their parts, but surprising leaps sustained by special
sub-conventions. Try explaining any of the following to a child or a foreigner:
self-published book, well-read man, graveyard shift, bald-faced lie, home in on, stand us in good
stead, life insurance, sea-change, doughnut holes, thought better of it. The
words and their dictionary definitions, in such cases, put you on the road to
meaning but do not take you very far.
A corollary of all this vagueness and apparent confusion is
that speech is no rote process, but endlessly personal and creative: less a
matter of finding the exactly right words, more of taking the words that are
handy and making them say what we want, with a celerity that makes Humpty-Dumpty seem basically right
after all. The lexicon and implicit grammar of the language provide tools and
raw material, but the rest is always up to you.
But the slack in the system, the incompleteness and
indeterminacy, means that the genie of sound has endless room and need to go
back to work. Since there will always be numerous ways to phrase any given
thing, the hunt for appealing sounds can go on endlessly. The babbling stage
never really ends, but becomes a powerful undercurrent in everyday usage, playing
against the need to select words for denotation. In the case of things that
must be said repeatedly, we regularly fuss and fiddle till we arrive at
something more or less musical. Consider:
drunk
as a skunk
pig
ignorant
bully
pulpit
Skunks, of course, do not drink, nor is their behavior
especially erratic. True, they resemble drunks in having a strong smell, and
one can argue for a certain kind of connotative precision here. Unquestionably,
though, the simile’s main raison d’etre
is the repetition of unk. Having been
said once, that coda is already handy, so to speak, and can be pronounced with
appreciably less effort than a new set of two or three phonemes. “Pig
ignorant,” likewise, has nothing to justify it on semantic grounds — swine are
actually rather intelligent, as barnyard animals go — but survives due to the
jingly doubling of “ig.” Why, though, is that phrase so much less current than
“drunk as a skunk”? Leaving aside the awkward dragooning of a noun to serve as
an adjective, the reasons are mainly musical. The doubling of “ig” is slightly
difficult to execute, forcing a pause as your voice finds its way back up from
the deep-in-your-throat “guh” to the
high-in-your-mouth “ihhh,” and then the rhythm trails off weakly: //uu. The
overall effect is that the rhyme has fired prematurely. By contrast the first,
more popular phrase skips along nicely, with the “kuh” opening pretty easily
into the “ahh,” helped along by an easier rhythm: /uu/. However barren
intellectually, drunk as a skunk can
be exclaimed with gusto, like any good slogan. Rhyme, that old recycling
project, lends it not just ease and economy, but emphasis and what in a poem we
would call structure, a kind of wholeness
or spatiality that makes it easier to construe and remember.
As for bully pulpit,
it shows an extreme case: something we say mostly for the jangly fun of
it, tolerating rank uncertainty as to meaning. When Theodore Roosevelt coined
the phrase, a rare feature of his preppy idiolect was that bully meant cool, rad,
bitching, phat, awesome, and so on. He meant only that the presidency was a
really bitchin' pulpit. But ever since then the phrase, phonologically perfect but
grammatically perplexing, has mostly evoked confused associations with bullying, as in the schoolyard. People
who use the expression are either itching for a chance to explain it, or intend
a sort of punning malapropism.
How common is this quest for euphony, this natural selection
and evolution of phrases on aural and aesthetic grounds? Surely it is almost
universal, if by “euphony” we mean not just superficial prettiness, but overall
ease and efficiency of articulation. Other things being equal — though granted,
they never quite are — we prefer economy to prolixity, easy transitions to
hard, regular rhythms to irregular, and above all, re-using the same sounds to
hunting up new. We English teachers like to speak of alliteration, assonance, and rhyme,
especially in poetry, as if they were magic adornments somehow added on
after the phrase was already (but how could it be?) complete. But what a
musical phrase really accomplishes, very generally speaking, is a reduction or
simplification to as few sounds as
possible, though ipso facto those few
are then recurrent and numerous. If substance is not too badly mangled, the
result is a special ease and rightness, a partial return of denotative speech to
the happy fluency of babble. There is nothing trivial about such
accomplishments; as every editor, engineer, or accountant knows, when you achieve
x more easily, you have energy left
over to accomplish y. To get a thing
said easily is a first step toward saying more,
and the aesthetics of speech, like other kinds of aesthetics, concerns long-range
efficiency quite as much as momentary pleasure.
xamples of the hunt for music are so numerous that the hard
thing is knowing where to begin. With simple initial alliterations[1] like rough and ready, chump change, tried and true, watching and waiting, or bad
to the bone? With terminal alliteration, as in weak link? With subtler assonance, as in Holy Moses, snow cone, itchy
fingers, bad actor, gay blade, long
haul, guy wire, all gone, saving
grace, right mind, or (in a
coinage from a women’s magazine) cleavage
wielding? With fully rhymed expressions like big rig, leaner and meaner,
boob tube, humdrum, fair share, or pooper scooper? With sneaky
muted half-rhymes like money hungry, mad as a hatter, heir apparent, or big picture?
With minimal pairs like tall tale, bee’s knees, sad sack, flip flop, and ship shape?
Or perhaps with a list of brand names,
selected for their mnemonic force after marathon meetings and extensive field
tests:
Coca-Cola
Roto-Rooter
Band-Aid (note
terminal alliteration and b/d
synergy)
Glad
Bags
Pizza
Hut (with a t lurking in the z)
Pepto-Bismol (mild assonance and p/b synergy,
but above all a forced, trochaic rhythm)
Or with the titles of books, movies, and TV shows:
Dialing for Dollars
Hogan’s Heroes
Beauty and the Beast
The Grapes of Wrath (an iambic beat and subtle gr/wr synergy)
Catch Twenty-Two (four ts in four syllables, and a total of
five aspirated consonants)
Or with proverbs and mnemonics, clinching their wisdom with
catchy rhythms and salient echoes:
First
deserve, and then desire.
A
stitch in time saves nine.
A
friend in need is a friend indeed.
A
miss is as good as a mile
Spare the rod and spoil the child.
Neurons that fire together, wire
together.
I
before E, except after C, or when sounded as A, as in neighbor and weigh
Such fish can be shot by the dozen
in any half-full barrel of English; so let’s hurry on to subtler examples.
Stare at an unobtrusive, sturdy little idiom like small talk, and you soon spot the quiet rhyme that makes it so serviceable.
Granted, the phrase is appealing on semantic grounds; the unexpected yoking of the
tangible and the intangible works like a metaphor, achieving a fine expressive
edge, and like any catchphrase this one performs a valuable service in reducing
a complex concept to shorthand. But that compression is largely arbitrary, a
matter of secondary convention attaching after the aural feat is done. The full
meaning — “inconsequential conversation,
made in a nervous or desultory manner, often in a nearly explicit effort to
evade larger issues” — arises only in the most preliminary sense from the
separate meanings of small and talk, which is why the phrase usually has
to be explained to foreigners and youngsters. (Someone guessing, as all
speakers must frequently do, might reasonably conclude that the phrase meant whispering, or a very short
conversation, or the opposite of big talk.)
One might say that small talk is a
word in its own right: a fetching sound first conjured up for its own sake, to
which meaning is then assigned, almost arbitrarily, in due course.
Or take politically
correct, a coinage that has been current for perhaps three decades. (Some
of us can remember the dark days when
we had to make do with far less snappy circumlocutions.) The soul of the
phrase, no doubt, is the dead-on irony with which it nails the contradiction between
generous political visions and the fussy enforcement protocols they generate. But
the rest of the success story has to do with sound: with the way the t and k (or “hard c”) of politically
are caught up and echoed in correct, the
way the p shares aspiration with the k (twice) and breath stoppage with both t and k, the
way the businesslike “short i”
repeats twice, rapid fire, then relaxes into the longer vowels of k’lee and ko, and the way the l flitters
through the first word, then returns in drag, as the r of correct.
Once you get started, it’s hard to stop. We would probably call
crossdressing something else if not
for the pleasant sibilance and the cr/dr synergy
of that expression. (There is transvestitism,
of course, but notice: three s’s, three
t’s.) Tie-dying might never have been the fad it was if the term were
not so deliciously convenient to say, with a mere flick of the tongue
distinguishing the first syllable from the second. A bounce house, hired out for street
fairs and birthday parties, has evolved rapidly from its forebear the inflatable house on the strength of
economy, near rhyme, and the long, spondaic, virtually onomatopoetic bounce
from one syllable to the next. Yet none of these expressions strikes us as
self-consciously poetical or literary; the music sits there quietly at its oar,
just doing its job.
A problem with the terms assonance
and alliteration, and with our
rough-and-ready habit of speaking of “s sounds,” “e sounds,” and so on, is that
orthography catches aural effects quite imperfectly. The two s’s in Says are actually different (though
similar) sounds; the o of trolley is less like the o of only
than the a of father. Slips on the ice is more alliterative
than it looks, if one allows a combination of initial and terminal sibilance. But
even a much more exacting code, like the International Phonemic Alphabet that
linguists use, would fail to catch many of the subtler dynamics of euphony; for
often what matters is not the echoing of whole phonemes, but of their “features”
or constituent qualities. Sounds that are similar but not identical can
reinforce each other, like the p and b of pop
bottle — or the p and the t, similar in a different way — an effect I have been calling
“synergy.” There are, in other words, all different degrees of alliteration and
assonance, rather few of which are honored in transcription
For a quick, rough example, take buzzsaw. Nothing in the transcription of the first syllable reveals
any repetition, but a single feature, voicing — that hum you make down in your
throat — carries through b, u, and zz, making the combination just a touch
easier to execute, if I have not had too much coffee this morning, than (let’s
say) bush or bus or puzz would be. Voicing then ceases with the s of saw,
but here two new features carry through, the tongue staying in the same
position and the breath making the same leaky hiss as we proceed from z to
s, two closely similar sounds of course, both classed as “alveolar
fricatives.” Notice again — this will be on the final — how moot the semantic aspect
of the word is. Wheelsaw, roundsaw, whinesaw, screamsaw, or beltsaw
would make as good logical sense, but those are unstable compounds and break
down with repeated pronunciation. Once we hit on buzzsaw, though, we like it and we keep it. (Circular saw is about equally current, but sibilance is a big part
of the reason.)
Alliteration, assonance, and their combination into rhyme seem
to be just the most obvious instances of a surprisingly persistent drive to
simplify and streamline speech, to optimize it for sayability, even at the cost
of some forcing of dictionary definitions. When the genie of sound is
especially busy and successful, the results can be uniquely memorable, as in
advertising jingles and nursery rhymes:
To
market, to market
To buy a fat pig;
Home
again, home again
Jiggety-jig.
To
market, to market
To buy a fat hog;
Home
again, home again
Jiggety-jog.
The meaning is almost negligible, but the two little verses
can be memorized on a single hearing. Try that with 30 words of prose. The secret
lies not just in the repetition of phonemes, or in the accentual meter
measuring out a steady two beats per line, or in the nearly verbatim
duplication of the first verse in the second, but also in the way different
phonemes resemble and reinforce one another. The whole “plosives” line on the
linguist’s sound chart gets a systematic workout, with b, p, t, k, and g all
salient and numerous, each with its little explosion of air. Not far away on
the chart, breathy aitches come up in “home” and “hog,” with fellow
“fricatives” j and f playing variations on the theme.
Bilabiality comes up in “buy,” “pig,” and “market,” nasality in “market” and
“again.” Voicing toggles on and off between the k of “market” and the g of
“pig,” and between initial consonants in “pig” and “buy.” “Jig” forms minimal
pairs with both “pig” and “jog,” and the shift from “jig” to “jog” features an
easy move from a high-front to a low-back vowel, in conformity with the
phonological rule that stipulates such sequences in terms like ping pong, flip-flop, and teeter-totter.
In such cases the urge to arrange sounds for their own sakes
gets the upper hand all but completely, and we readily class the poem as
“nonsense,” especially in the “jiggety” lines, where repetition reigns supreme
and denotation goes off the radar. Hearing the babbling voice of infancy, devoted
to sound-practice rather than meaning, we instinctively categorize the poem as
for, and perhaps by, children.
But adult language, too, offers many instances of aural
playfulness. In her essay “Ablaut Schmablaut,” my friend Valerie Collins
discusses a wide range of “reduplicative” expressions in English — words or
phrases like helter-skelter, mumbo-jumbo, criss-cross, mish-mash, razzle-dazzle, yo-yo, tut-tut, higgledy-piggledy, and many more. She
notes that such “fun-motivated expressions” feel playful not because they ignore
the usual rules but more because they rehearse the rules to excess, beyond what
is really needed to form distinct, vivid words. “The central impulse for making
these reduplicatives is form . . . the sound pattern is the primary
motivation for the coinage.”
Reminiscent of crib talk, such
terms have an archaic feel; their odd copiousness makes them seem vaguely
unwordlike, not quite right for serious discourse. In principle if not quite in
fact, the reduplicatives are relics of early play, left lying around the
language the way abandoned toys might lie around a childhood home one still
occupies as an adult. We don’t really need them, but for affection’s sake we grandfather
them into the adult lexicon, granting a special exemption to the criterion of
economy.
In any case, the drive for economy, which seems to yield
alliterative expressions by way of sound reduction, is just one side of the
coin; the other is the need for hearable distinctions, and at some point, it
seems, variety rather than sameness becomes valuable for its own sake. Poetry
texts nearly always include a caveat to the effect that alliteration can be
overdone, accompanied by cautionary specimens:
There ought
to be capital punishment for cars
that run
over rabbits and drive into dogs
and commit
the unspeakable, unpardonable crime
of killing
a kitty cat still in his prime. [2]
The little
toy dog is covered with dust
But sturdy
and staunch he stands. [3]
Horrendous
horrors haunted Helen’s happiness. [4]
The lurch into bathos is unmistakable, but to my mind rather
hard to account for. Why should so many other alliterative expressions flow by
so unobtrusively and usefully in the stream of language, while these
nonce-works annoy and perplex? The sheer concentration of repeated sounds
does not seem to account for it, as there are specimens of great poetry where
the frequency is as high. John Frederick Nims offers an example from Coleridge:
The fair breeze blew, the white
foam flew,
The furrow followed free.
We were the first that ever burst
Into that silent sea. [5]
In the cautionary examples, it can be hard to separate the
effect of alliteration itself from other musical problems, and from the general
thick-headedness. Still it is noticeable, and curious, that bad alliteration is
a vaguely quantitative proposition. Two, three, or four bs are fine, but the fifth turns all gain to loss.
One of my favorite rock lyrics of all time, from “Werewolves
of London” by Warren Zevon, skates right up to this line between brilliance and
bathos: “A little old lady
got mutilated late last night.” Say it once, even without the tune, and you
will say it a thousand times, but be sick of it and cursing my name long before
you quit. Artful or awful? You decide.
My guess is that the recoil effect concerns not just ease of
articulation, but another aspect of alliteration: structure. In euphonious
language, repeated sounds help give a phrase what literary criticism used to
call a “spatial dimension”: they frame it, solidify it, give it a certain wholeness
and autonomy within the stream of speech. But too much spatiality leaves us
spaced out and confused; the phrase we are making looms up too stark and static
in the mind’s eye, and we become uncomfortably self-conscious of language as
mere noise. Articulation takes the foreground, while reference or meaning falls
away. The effect is especially noticeable in tongue twisters: “Rubber baby
buggy bumpers” is not so much an enunciation problem as a processing problem:
on the second or third go-round, your inner genie loses its place, unable to
tell whether this particular b is the
one in “rubber” or in “buggy.” Bad alliteration can be partly a numerical issue
because it is a matter of reaching (or not) the limits of what your
sound-processing systems can handle.
And of course there are myriad other complications of and
exceptions to the principle of euphony. The aesthetic impulse sometimes
registers an unexpected preference for hard shifts over easy ones, unlike
sounds over like, arrhythmia over a healthy beat, and even prolixity over
economy. Sometimes difficulty is the point. The extra effort needed to spit out
jackbooted thug or jumpin’ Jehosaphat is part of the
meaning, a case of expressive form. Horse
feces is rather easier to say than bullshit, even with the additional
syllable, but fails to enchant partly because the whispery sibilants and
squeaky vowels weaken what should be a pugnacious sentiment. A phrase like spontaneous combustion, fun to say but
somehow a bit too fancy and self-admiring, features plenty of synergy but fails
the test of economy. Mouth-filling PC terms like African American, developmentally
disabled, and chairperson deliberately
overburden the tongue and ear, the better to express (or flaunt) solemnity and
contrition. Bad academic prose employs a whole arsenal of techniques aimed at
securing a special, rather artful kind of unintelligibility that listeners will
blame on themselves rather than the professor. Indeed, jargon of every kind
makes a fetish of ugliness, by way of claiming special authority for the guild
that uses it.
Thank goodness, then, for the poet’s “idiot fiddling with
the sounds of things,” as Howard Nemerov called it. Thank goodness for the playfulness
that preserves “the health of the language” if anything does. Earnest attempts
to zero in on meaning have a way of arriving at preposterous contrivances — shoulder prosthesis program evaluation officer protocol — while mere play quietly produces
graceful, useful coinages that enrich everyone’s ability to make sense.
Are we, then, the slaves of sound? Does the coincidence of like
phonemes turn thought into channels it would otherwise avoid? Sometimes, no
doubt, but not often. Most of us, most of the time, are intent on meanings that
would have much the same shape and urgency no matter what phrasing we found. We
are more apt to warp our phrases to fit them
than vice versa, and the natural vagueness of language, its
Humptydumptismo, makes this perfectly permissible up to a point. In any case
the quest for euphony is more the culture’s than our own: not so much a matter
of inventing on the fly as of sensibly relying, short of cliché, on the fruits
of past creativity. The catchphrases lie there ready-made (indispensably so,
for all of George Orwell’s misguided strictures against using them), and you
choose the ones you need while letting the others alone. What makes it all
possible is the great elasticity of words, the readiness with which they extend
themselves by means of metaphor or irony, submitting to provisional definitions
but then returning unscathed to their dictionary senses. We teachers urge
students to use words “correctly” and “carefully,” and there is a good reason
for this: we want to get them into the ballpark. But once inside, they can
choose from thousands of seats.
Terrible twos alliterates
and is therefore a catchphrase in English, while terrible threes does not and is not. I suppose it’s possible that a
parent in Chicago might therefore be more disposed to expect early tantrums than
one in Barcelona. But to think that the phrase somehow prevents her from recognizing misbehavior in three year-olds, and
from describing it in any of a thousand ways (e.g., mean threes) is plain silly. A few years back the Atlantic Monthly ran an article on “road
rage,” which at that point the blatherskites on TV had been talking up into an
“alarming trend.” The gist of the article, supported by various studies and
statistics, was that there was no trend. What had happened was simply that the
phrase had been coined and people liked to repeat it, creating the impression
that the thing itself was on the increase. The episode shows our occasional tendency
to be misled by the fun of a nifty phrase; but the existence of the article
proves how readily thought recovers from such misadventures. The resources of English
or any language are such that the thinkable is always sayable, but not dictated
by it. If you know what you mean, you find a way to get it across.
Notes
[1]
The term is commonly restricted to the
recurrence of consonant sounds at the beginnings of words, and somewhat less
commonly extends to echoes at the beginnings of syllables even in mid-word. I
use it more broadly still, to refer to consonant echoes in any position at all.
That is, I include consonance,
sometimes calling it terminal
alliteration, in order to have a generic term.
[2] McKuen, Rod. “Thoughts on Capital
Punishment.” In X.J. Kennedy, An
Introduction to Poetry, 7th ed. Glenview, Illinois: Scott
Foresman, 1990: 252.
[4]
Meyer, Michael, Poetry: An Introduction. Boston:
St. Martin’s Press, 1995: 138. A made-up example, surely.
[5]
Nims, John Frederick: Western Wind: An
Introduction to Poetry. New York: Random House, 1974: 189. The poem is Rime of the Ancient Mariner by Samuel
Taylor Coleridge.
