American Language (84 page)

“The three great causes of change in language,” says A. H. Sayce, “may be briefly described as (1) imitation or analogy, (2) a wish to be clear and emphatic, and (3) laziness. Indeed, if we choose to go deep enough we might reduce all three causes to the general one of laziness, since it is easier to imitate than to say something new.”
⁸⁷
This tendency to take well-worn paths, paradoxically enough, seems to be responsible both for the transfer of verbs from the strong to the weak declension, and for the transfer of certain others from the weak to the strong. A verb in everyday use tends almost inevitably to pull less familiar verbs with it, whether it be strong or weak. Thus,
fed
as the preterite of
to feed
and
led
as the preterite of
to lead
eased the way in the American vulgate for
pled
as the preterite of
to plead
; and
rung
as plainly performed the same office for
brung
, and
drove
for
dove
and
hove
, and
stole
for
dole
, and
won
for
skun
. Contrariwise, the same combination of laziness and imitativeness worked toward the regularization of certain verbs that were historically irregular. One sees the antagonistic pull of the two influences in the case of verbs ending in
-ow
. The analogy of
knew
and
grew
suggests
snew
as the preterite of
to snow
, and it is sometimes encountered in the American vulgate. But meanwhile
knew
and
grew
have been themselves succumbing to the greater regularity of
knowed
and
growed
. So
snew
, losing support, grows rare and is in palpable decay, but
knowed
and
growed
show great vigor, as do many of their analogues. The substitution of
heerd
for
heard
also presents a case of logic and convenience supporting analogy. The form is suggested by
feared, cheered, cleared
, etc., but its main advantage lies in the fact that it gets rid of a vowel change, always an impediment to easy speech.

Some of the verbs of the vulgate show the end-products of other language movements that go back to a very early period. There is, for example, the disappearance of the final
t
in such words as
crep
,
slep, lep, swep
and
wep
. Most of these, in Old English, were strong verbs. The preterite of
to sleep
(slœpan), for example, was
slep
, and of
to weep
was
weop
. But in the course of time both
to sleep
and
to weep
acquired weak preterite endings, the first becoming
slœpte
and the second
wepte
. This weak conjugation, in most cases, was itself degenerated. Originally, the inflectional suffix had been
-de
or
-ede
and in some cases
-ode
, and the vowels were always pronounced. The wearing-down process that set in in the Twelfth Century disposed of the final
e
, but in certain words the other vowel survived for a good while, and we still observe it in such archaisms as
learnéd
and
belovéd
. Finally, however, it became silent in other preterites, and
loved
, for example, began to be pronounced (and often written) as a word of one syllable:
lov’d
.
⁸⁸
This final
d
-sound now fell upon difficulties of its own. After certain consonants it was hard to pronounce clearly, and so the sonant was changed into the easier surd, and such words as
pushed
and
clipped
became, in ordinary conversation,
pusht
and
clipt
. In other verbs, the
-t
(or
-te
) ending had come in long before, and when the final
e
was dropped only their stem vowels needed to be changed. Thus arose such forms as
slept
. In vulgar American another step is taken, and the suffix is dropped altogether. Thus, by a circuitous route, verbs originally strong, and for many centuries hovering between the two conjugations, have eventually become strong again.

The case of
helt
is probably an example of change by false analogy. During the Thirteenth Century, according to Sweet
⁸⁹
“
d
was changed to
t
in the weak preterites of verbs [ending] in
rd, ld, nd.
” Before that time the preterite of
sende
(
send
) had been
sende
; now it became
sente
. It survives in our modern
sent
, and the same process is also revealed in
built, girt, lent, rent
and
bent
. The popular speech, disregarding the fact that
to hold
is a strong verb, arrives at
helt
by imitation.
⁹⁰
In the case of
tole
, which I almost always hear in place of
told
, there is a leaping of steps. The
d
is got rid of by assimilation with
l
and without any transitional use of
t
. So also, perhaps, in
swole
, which is fast displacing
swelled. Attackted
and
drownded
seem to be examples of an effort to dispose of harsh combinations by a contrary process. Both are old in English.
Boughten
and
dreampt
present greater difficulties. Lounsbury says that
boughten
probably originated in the Northern (
i.e.
, Lowland Scots) dialect of English,
“which … inclined to retain the full form of the past participle,” and even to add its termination “to words to which it did not properly belong.”
⁹¹
The
p
-sound in
drempt
follows a tendency that is also seen in such pronunciations as warm(
p
)
th
,
com
(
p
)
fort
and
some
(
p
)
thing
, and that has actually inserted a
p
in
Thompson
(Tom’s son). The general movement toward regularization is well exhibited by the new verbs that come into the language constantly. Practically all of them show the weak conjugation, for example,
to broadcast
.
⁹²
Even when a compound has as its last member a verb ordinarily strong, it is often weak itself. Thus the preterite of
to joy-ride
is not
joy-rode
, nor even
joy-ridden
, but, unless my ears fail me,
joy-rided
. And thus
bust
, from
burst
, is regular and its usual preterite is
busted
, though
burst
is irregular and its preterite is the verb itself unchanged. The same tendency toward regularity is shown by the verbs of the
kneel
class. They are irregular in English, but tend to become regular in colloquial American. Thus the preterite of
to kneel
, despite the example of
to sleep
and its analogues, is not
knel
’, nor even
knelt
, but
kneeled
. I have even heard
feeled
as the preterite of
to feel
, as in “I
feeled
my way,” though here
felt
still persists.
To spread
also tends to become weak, as in “He
spreaded
a piece of bread.” And
to peep
remains so, despite the example of
to leap
. The confusion between the inflections of
to lie
and
to lay
extends to the higher reaches of spoken American, and so does that between
lend
and
loan
. In the vulgate the proper inflections of
to lend
are often given to
to lean
, and so
leaned
becomes
lent
, as in “I
lent
on the counter.” In the same way
to set
has almost completely superseded
to sit
, and the preterite of the former,
set
, is used in place of
sat
. But the perfect participle (which is also the disused preterite) of
to sit
has survived, as in “I have
sat
there.”
To speed
and
to shoe
have become regular, not only because of the general tendency toward the weak conjugation, but also for logical reasons. The prevalence of speed contests of various sorts, always to the intense interest of the proletariat, has brought such words as
speeder, speeding, speed-mania, speed-maniac
and
speed-limit
into
daily use, and
speeded
harmonizes with them better than the irregular
sped
. The American’s misuse of
to learn
for
to teach
is common to most of the English dialects. More peculiar to his speech is the use of
to leave
for
to let
. Charters records it in “Washington
left
them have it,” and there are many examples of it in Lardner.

In studying the American verb, of course, it is necessary to remember always, as Menner reminds us, that it is in a state of transition, and that in many cases the manner of using it is not yet fixed. “The history of language,” says Lounsbury, “when looked at from the purely grammatical point of view, is little else than the history of corruptions.” What we have before us is a series of corruptions in active process, and while some of them have gone very far, others are just beginning. Thus it is not uncommon to find corrupt forms side by side with orthodox forms, or even two corrupt forms battling with each other. Lardner, in the case of
to throw
, hears “if he had
throwed
”; my own observation is that
threw
is more often used in that situation. Again, he uses “the rottenest I ever seen
gave
”; my own belief is that
give
is far more commonly used. The conjugation of
to give
, however, is yet very uncertain, and so Lardner may report accurately. I have heard “I
given
” and “I would of
gave
,” but “I
give
” seems to be prevailing, and “I would of
give
” with it, thus reducing
to give
to one invariable form, like those of
to cut, to hit, to put, to cost, to hurt
and
to spit
. My table of verbs shows various other uncertainties and confusions. The preterite of
to blow
may be
blowed, blew
or
blown
, and that of
to drink
oscillates between
drank
and
drunk
, and that of
to fall
is still usually
fell
, though
fallen
has appeared, and that of
to shake
may be either
shaken
or
shuck
. The conjugation of
to win
is yet far from fixed. The correct English preterite,
won
, is still in use, but against it are arrayed
wan
and
winned
, and Lardner, as I have noted, believed that the plain form of the present would eventually oust all of them.
Wan
seems to show some kinship, by ignorant analogy, with
ran
and
began
. It is often used as the perfect participle, as in “I have
wan
$4.” This uncertainty shows itself in many of the communications that I have received since my last edition was published. Practically every one of my conjugations has been questioned by at least one correspondent; nevertheless, the weight of observation has supported all save a few of them, and I have not made many changes.

The misuse of the perfect participle for the preterite, so common
in vulgar American, is also common in many other dialects of English. It has been going on for a long time, and in American, the most vigorous and advanced of all the dialects of the language, it is particularly well marked. Menner believes that it originated, at least as to some of the verbs, in the decay of the auxiliary
have
in the present perfect. The omission of the auxiliary, he says,