American Dialect Society

Jewish Dialect and New York Dialect

Author(s): C. K. Thomas
Source: American Speech, Vol. 7, No. 5 (Jun., 1932), pp. 321-326
Published by: Duke University Press
Stable URL: http://www.jstor.org/stable/452953
Accessed: 20-02-2016 03:43 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.

Duke University Press and American Dialect Society are collaborating with JSTOR to digitize, preserve and extend access to
American Speech.

http://www.jstor.org

This content downloaded from 160.39.5.105 on Sat, 20 Feb 2016 03:43:04 UTC
All use subject to JSTOR Terms and Conditions

VOLUME VII

NUMBER 5

domerican

Speech

JUNE - 1932

JEWISH DIALECT AND NEW YORK DIALECT

C. K. THOMAS
Cornell University
URINGthe past four years I have worked with some hundreds of

university students in an attempt to improve the quality of

their speech. A fair proportion of these students have been
Jews from New York City and its suburbs. Their social and scholastic
levels are about the same as those of other New Yorkers, but their
speech is distinctly inferior, and this inferiority raised the question
whether there might be a clearly defined dialect which was characteristic of New York Jews. The students with whom I have worked do
not, of course, constitute a true cross-section of either New York or
Jewish speech; such a cross-section would have to be obtained in New
York itself. Those who can afford to travel 250 miles for their education represent, on the average, a higher social and economic level than
those who stay at home and who are able, in many cases, to earn a
larger part of their expenses than is possible in a small town. Because
of this higher level, and because few of the New York Jews at Cornell
speak any language but English, their dialect is by no means as extreme
as that of the peripatetic Mr. Klein so carefully studied by Miss
Benardete,1 or even as extreme as that of the general run of Jewish
undergraduates in the New York City colleges. Many of them, however, have complicated their speech problem with tricks acquired in the
elocution schools that are at present so popular among the higher
class Jewish families of New York. Traditional Jewish and traditional
New York pronunciations alike are in some cases conspicuously absent.
Moreover, most of the students with whom I am familiar are to some
extent conscious of their speech, for the greater number of them are sent
1 Dolores Benardete, "Immigrant Speech-Austrian-Jewish Style,"
SPEECH,

AMERICAN

October, 1929.
321

This content downloaded from 160.39.5.105 on Sat, 20 Feb 2016 03:43:04 UTC
All use subject to JSTOR Terms and Conditions

American Speech

322

to me from courses in public speaking and dramatics. All of these

factors complicate the problem of analysis, and make the results less
conclusive.
For purposes of comparison I divided my students into three groups:
(1) Jews from New York City and its suburbs, (2) Gentiles from the
same area, (3) Jews who had lived all their lives at a distance from New
York City. After discarding all doubtful cases, namely, those of
uncertain racial origin, those who had lived both in New York City and
elsewhere, and those who lived in the typically Jewish summer resorts
of the Catskills, I was left with the records of 112 students, of whom 75
were New York Jews, 19 were New York Gentiles, and 18 were Jews
from other parts of the country. Thus approximately 67 percent of
the total was in group 1, 17 percent in group 2, and 16 percent in group
3. A normal distribution of dialectal peculiarities, or errors, would
therefore result in the same percentages, but when the errors had been
classified it was found that, out of a total of 673, the New York Jews
had made 522, the New York Gentiles 71, and the Jews from other
parts of the country 80. In other words, group 1 made 78 percent of
the errors, group 2 made 10 percent, and group 3 made 12 percent.
Thus, in comparison with an average distribution, the speech of group 1
was distinctly inferior to that of the other two groups, as the following
summary shows:

Group
1
A Number of cases in each group.........
B Percentage of cases...................
C Number of errors in each group........
D Percentage of errors..................
E Percentage above or below average dis.......
tribution2.............

75
67
522
78
+16

19
17
71
10

18
16
80
12

-38

Total
112
100
673
100

-26

In considering the distribution of particular errors among the three

groups, one must refer to line D in the above table as a basis for comparison. If the percentages for the particular error do not vary greatly
from those of line D it is obvious that they give no information regard2 These figures represent the variation from 100 of the quotients obtained by
dividing the figures in line D by the correspondingfigures in line B; in all calculaions the percentages were carried to two extra decimals.

This content downloaded from 160.39.5.105 on Sat, 20 Feb 2016 03:43:04 UTC
All use subject to JSTOR Terms and Conditions

Jewish Dialect

323

ing the source of the error; but if the percentage for group 1, consisting
of New York Jews, is higher than 78, that of group 2, consisting of New
York Gentiles, higher than 10, and that of group 3, consisting of Jews
from other localities, lower than 12, the distribution creates a strong
presumption that the error in question is local rather than racial, for
only the groups which include New Yorkers show a higher percentage
than the average distribution. Similarly, if the percentages for another
error are above the average for groups 1 and 3, and below for group 2,
the distribution creates a strong presumption that this is a racial,
rather than a local, error, for only the groups which include Jews show
higher percentages than the average. In many cases, of course, there
are neither sufficiently large numbers of instances of the error nor
sufficiently great variations from the average to warrant any definite
conclusion; in other cases, which are listed below, definite conclusions
are inescapable.
The most frequent error among these students was the dentalizing
of the alveolar consonants [t, d, n, 1, s, z];"the error consists in making
the characteristic consonantal obstruction between the tongue and teeth
instead of between the tongue and gum ridge. The acoustic effect of
this misplacement is least noticeable for [n, 1]; for [s, z] it suggests a
slight lisp; [t, d] sound overexplosive and slightly higher in pitch. It is
most noticeable when several alveolar consonants appear in the same
word, as in dental and slant. The distribution of this error clearly
indicates that it is Jewish in origin: group 1 is 10 percent above, and
group 3 is 5 percent below, the average distribution of line D; but group
2, the Gentile group, is 66 percent below the average. In short, the
Gentile group is remarkably free from this error, including only 8
instances out of a total of 224. The most frequently dentalized of
these consonants is [1], and here the distribution is even more clearly
Jewish: 10 percent above the average for group 1, 5 percent above for
group 3, and 80 percent below for group 2. The cause of this error,
whether a survival from Yiddish, German, or Slavic linguistic habit or
otherwise, is not within the scope of this paper.
Closely associated with dentalization is the overaspiration of [t]
after [n] or [1], particularly at the beginning of an unstressed syllable
or at the end of a word, as in winter, wilted, went, and wilt. Here the
percentages are inconclusive, but it seems likely that this error is also
Jewish.
Letters in square brackets are phonetic characters, which refer to sounds;
those in quotation marks refer to spellings.

This content downloaded from 160.39.5.105 on Sat, 20 Feb 2016 03:43:04 UTC
All use subject to JSTOR Terms and Conditions

324

American Speech

Difficulty with [s] seems to be characteristic of Jewish speech. This

is in part owing to the habit of dentalizing, but in addition there are
other errors: exaggerated hissing, substitution of voiceless [1]and "th"
[6] as in thin, and occasionally "sh" [f] as in she. There are 35 instances
of these variations, of which only 3 are Gentile. Group 3, consisting of
Jews who do not live in New York City, has the greatest difficulty with
this sound, being 68 percent above the average distribution.
Another clearly Jewish error is the substitution of [rig]for [i], so as,
for instance, to make singer a rhyme for finger. There are 35 instances
of this error in the three groups, and only one of them is Gentile.
Furthermore, according to these figures, the Jews from New York do
not make this error quite as persistently as do those from other localities. The substitution of [ijk] for [rJ],so as, for instance, to make sing
and sink identical in sound, appears only 8 times. Though this is the
traditional form of the error, perhaps because it can be represented
more easily in the conventional alphabet, I do not believe it to be
nearly as common as [rig]. Once in a great while the glottic stop is
added instead of either [k] or [g].
Loss of the distinction between the voiced [w] and the voiceless
"wh" [Ml,so as, for instance, to make witch and which identical, is
quite common in both New York groups, but less common among the
Jews from other localities. The distribution is 4 percent above the
average from group 1, 60 percent above for group 2, and 29 percent
below for group 3. In other words, the Jews from outside of New
York have least trouble with the voiceless [M],and this bears out the
traditional notion that, although this error is by no means confined
to New York, it is there most conspicuous and prevalent.
Similarly, the addition of an [r] to such words as idea and law,
especially when the following word begins with a vowel, is, at least for
these three groups, a New York characteristic, for none of group 3
added the [r], and group 2, the Gentile group, added it more consistently
than group 1. The error is not, of course, limited to New York City,
but is also encountered in New England.
Errors in vowels and diphthongs are, with some doubtful exceptions, New Yorkese rather than Jewish. The vowel [o(v)] is distorted
into an exaggerated diphthong which can best be indicated as [ev] or
[ev], as in the pronunciation [nevt] for note [no(v)t]. This is similar to
the extreme pronunciation of Oxford, though it is drawled to a greater
length in New York. Group 1 is 2 percent above the average for this
error, group 2 is 31 percent above, but group 3 is 42 percent below.

This content downloaded from 160.39.5.105 on Sat, 20 Feb 2016 03:43:04 UTC
All use subject to JSTOR Terms and Conditions

Jewish Dialect

325

It should be noted that the Gentile group has the most trouble with
this sound.
Another characteristic error is the substitution of the compromise
[a] for the flat [m] in such words as land, man, and bad. This differs
from the New England use of [a] and the Southern British use of [a] in
such words as path, dance, and laugh, and is more like certain Scotch
and Irish dialects. A possible explanation may lie in the concerted
efforts now being made in New York to teach the "broad a" [a] of
"world standard" English. One who acquires this "broad a," or
even the compromise [a], some years after learning to speak English is
likely to use it in the wrong words, and at the same time to get the
impression that the "flat a" [oe]is a disreputable sound, to be avoided
whenever possible. At any rate, group 3 is least susceptible to the
error, and group 1, which has had the greatest amount of elocutionary
training, the most susceptible.
Substitution of [yev]for [av] in such words as now, out, and power
appears not to be a Jewish error. Group 1 is 3 percent below average,
group 3 is 5 percent below, and group 2, the Gentile group, is 29 percent
above. This error is characteristic of the South and of rural New
England as well as of New York, and its significance in this study is
doubtful.
The change of the diphthong in my, fine, and light from [ai] to [aI], or
to an even more retracted form, appears to be a New York characteristic, though more data will be required for certainty. In its most
characteristic form the distortion resembles the German variety of the
diphthong more closely than anything else. Group 1 is 12 percent
above the average; group 2 is 10 percent below; group 3, however,
includes only one instance of the error.
Statistical figures on vocal quality are much less reliable, as the
qualities themselves are so variable. In general, however, indistinctness resulting from inactivity of the lips appears to be a New York
characteristic, drawl is more common among the Jews, and "throatiness" exclusively Jewish. Nasality is common, and not limited to
either group.
So far, then, as can be learned from the data of this study, the New
York Jew dentalizes the alveolar consonants, overaspirates [t], has
and has a drawling, throaty vocal
various difficulties with [s] and
[ra],
quality because he is Jewish; on the other hand, he uses the voiced
[w] for the voiceless [&], substitutes [ev] for [o(v)], [a] for [em],[mev]for
[av], and [ai] for [ai] adds, the intrusive [r], and uses his lips insufficiently

This content downloaded from 160.39.5.105 on Sat, 20 Feb 2016 03:43:04 UTC
All use subject to JSTOR Terms and Conditions

326

American Speech

because he is a New Yorker. Obviously these conclusions are tentative, and much more data will be required before any conclusions
approaching finality can be reached; but it seems evident, nevertheless,
that a good bit of what passes popularly for Jewish dialect is really New
York dialect, and that details which pass unnoticed in Gentile speech
are more apt to be noticed in Jewish speech because of the lower
quality resulting from the mixture of errors from local and racial
sources.

This content downloaded from 160.39.5.105 on Sat, 20 Feb 2016 03:43:04 UTC
All use subject to JSTOR Terms and Conditions