FACTS ABOUT
CURRENT ENGLISH USAGE
THE NATIONAL COUNCIL OF TEACHERS OF ENGLISH
211 WEST SIXTY-EIGHTH STREET, CHICAGO, ILLINOIS
Officers for 1938
President, marquis e. shattuck ,
1st Vice-President, essie chamberlain
2nd Vice-President, e. a. cross
Secretary-Treasurer, w. wilbur hatfield
Executive Committee for 1938
DORA V. SMITH
CHARLES SWAIN THOMAS
HOLLAND D. ROBERTS
And the officers of the Council
PUBLICATIONS
Issued by special committees of the National Council of Teachers
of English under the direction of the Publications Committee:
Stella S. Center John J. DeBoer Homer A. Watt
Neal M. Cross Holland D. Roberts, Chairman
Reading for Fun (elementary book list)
. Compilation directed by Eloise Ramsey
Books for Home Reading (for High Schools)
Leisure Reading (for Grades Seven, Eight, and Nine)
Compilations directed by Stella S. Center and Max J. Herzberg, Com¬
mittee Co-Chairmen
Good Reading (for Colleges)
Compilation directed by Atwood H. Townsend, Committee Chairman
Current English Usage, Sterling Andrus Leonard, Committee Chairman
Guide to Play Selection, Milton Smith, Committee Chairman
The Teaching of College English, Oscar James Campbell, Committee Chair¬
man
Photoplay Appreciation in American High Schools, William Lewm, Com¬
mittee Chairman
An Experience Curriculum in English, W, Wilbur Hatfield, Committee
Chairman
A Correlated Curriculum, Ruth Mary Weeks, Committee Chairman
Teaching High School Students to Read, Stella S. Center and Gladys L.
Persons
Film and School, Helen Rand and Richard Lewis
Facts About Current English Usage, Albert H. Marckwardt and Fred G.
Walcott
Official Organ
The English Journal: College and High School Editions
Editor, W. Wilbur Hatfield, 211 West Sixty-Eighth Street, Chicago, Illinois
FACTS ABOUT
CURRENT ENGLISH USAGE
BY
ALBERT H. MARCKWARDT
AND
FRED G. WALCOTT
INCLUDING A DISCUSSION OF CURRENT USAGE
IN GRAMMAR FROM "CURRENT ENGLISH USAGE”
BY
STERLING A. LEONARD
A PUBLICATION OF
THE NATIONAL COUNCIL OF TEACHERS OF ENGLISH
D. APPLETON-CENTURY COMPANY
INCORPORATED
NE'W YORK
LONDON
COPYRIGHT, 1938, BY
THE NATIONAL COUNCIL OF TEACHERS OF ENGLISH
All rights reserved. This hook, or parts
thereof, must not be reproduced in any
form without permission of the publisher.
3118
PRINTED IN THE UNITED STATES OF AMERICA
PREFACE
This study, which the Publications Committee of the Na¬
tional Council of Teachers of English has generously consented
to make available in printed form, had its inception in two
papers presented before the Current Language Problems sec¬
tion of the National Council of Teachers of English. In 1933,
the year after the appearance of Current English Usage ,
Mr. Walcott read a paper entitled, “Some Practical Aspects of
the Leonard Monograph 55 at the Council meeting held in
Detroit. Mr. Marckwardt's paper was read at the Buffalo
meeting in 1937. Both papers had the same end in view and
employed the same technique. The authors collected evidence
as to the usage of certain expressions for the purpose of com¬
paring this evidence with the collected opinion in the Leonard
report about the same expressions.
Thus it seemed feasible to combine the two papers into a
single comprehensive study which would present the factual
evidence concerning all, or nearly all, of the expressions sur¬
veyed in Current English Usage . It is this combination which
is presented here. Mr. Walcott is responsible for the work on
those items which both groups of judges voted “established. 55
The items which were voted “established 55 by the linguists and
“disputable 55 by the per-capita vote of all the judges were
studied independently by both collaborators, the results being
combined in this monograph. The work on the remaining “dis¬
putable 55 and the “illiterate 55 items was done by Mr. Marck-
wardt.
To help those readers who are not familiar with the original
Leonard report, and to furnish a convenience for those who
may wish to compare Leonard's findings with ours, the gram¬
mar portion of the Leonard study has been appended to this
monograph.
A. H. M.
F. W.
V
CONTENTS
PAGE
Preface . v
I. The Leonard Study on Usage.1
The Technique of the Leonard Study.2
Presentation of Results .4
Grammatical Usages as Ranked by Linguists ... 4
II. The Purpose and Method of This Study .... 13
The Method in This Study.15
Results of the Investigation.17
III. The “Established” Usages.22
IV. The “Disputable” Usages..33
V. The “Illiterate” Usages.52
VI. Conclusion.61
CURRENT USAGE IN GRAMMAR
Reprinted from
Current English Usage by Sterling A. Leonard
1. Introduction.65
II. Judges’ Discussion of Specific Items of Usage . . 69
Nouns. . 70
Pronouns.71
Verbs..82
Adjectives.97
Articles.98
Adverbs.. . 98
vii
CONTENTS
viii
PAGE
Comparison. 105
Prepositions. ......107
Conjunctions and Conjunctive Adverbs.HI
Sentence Structure . H4
Barbarisms and Improprieties.116
Idioms and CoI!c^vdaI:sT.s.117
Changes in Definition and Use of Words.122
Debated Spellings '.HO
III. Practical Conclusions as to Grammar and Usage . . 132
Index .139
FACTS ABOUT
CURRENT ENGLISH USAGE
THE LEONARD STUDY ON USAGE
It was most appropriate that the first of the English Mono¬
graphs sponsored by the National Council of Teachers of Eng¬
lish should have dealt with the very fundamental problem of
language. This monograph. Current English Usage, is the result
of a study initiated by the late Professor Sterling A. Leonard.
It was completed after his untimely death by a committee of
the National Council and published in 1932. Appearing at a
time when the researches of Leonard, Fries, and Pooley, to
mention only a few, had established the unreliable character
of the linguistic judgments in many school textbooks, this study
sought to provide the classroom teacher with accurate and re¬
liable information concerning English usage. The constant in¬
terest of the teaching profession in research of such a practical
and helpful character is evidenced by the fact that the second
printing of Current English k Usage is now exhausted.
Although the appreciation of the teachers for this work has
been manifest, there are two other aspects of its reception that
deserve comment here. It should be recalled, first of all, that
Current English Usage was the subject of much adverse jour¬
nalistic comment both upon its initial appearance and when the
second printing was issued. In general, these criticisms were
either flippant or indignant; they seized upon what seemed to
be some of the most startling of the findings and used them as a
point of departure to predict the disintegration of the English
language or to question the sanity of the authors. Almost without
i
2 FACTS ABOUT CURRENT ENGLISH USAGE
exception the criticisms were neither penetrating nor construc¬
tive.
On the other hand, language scholars and language historians,
the individuals best qualified not only to make an intelligent ap¬
praisal of the results of the Leonard study but to pursue some
of the stimulating problems that were raised in it, have given it
little notice. During the five years in which Current English
Usage has been available to them, there have been few studies
based upon it, and these deal with the pedagogical rather than
the linguistic implications of the study. It is, in part at least, the
purpose of this monograph to demonstrate the possibilities of
linguistic analysis of the Current English Usage results.
THE TECHNIQUE OF THE LEONARD STUDY
Before proceeding, it will be pertinent, however, to recall a
few of the details concerning the development of this study, the
method pursued by Leonard and his associates, and the results
which they presented. The study appears in what may be called
its initiatory stage in an article by Leonard and Moffett, entitled
“Current Definitions of Levels in English Usage,” and published
in the English Journal for May, 1927 (pages 345-359). The pur¬
pose of the study is set forth in the following statement which
appears at the outset of the articles
This study was an attempt to find out what various judges have
observed about the actual use or non-use by cultivated persons of a
large number of expressions usually condemned in English textbooks
and classes.
The monograph Current English Usage, which appeared five
years later, was an enlargement of this earlier project, employing
the same method and having the same purpose in view. The first
significant fact to remember, then, is that Current English Usage
deals primarily not with usage itself but with opinion about the
THE LEONARD STUDY ON USAGE
3
usage of words and expressions usually questioned or condemned
in grammars and handbooks.
The following technique was employed in securing this “con¬
sensus of expert opinion”: a list of 230 expressions “of whose
standing there might be some question” 1 was submitted to a
group of 229 judges, composed of thirty linguistic specialists, an
equal number of editors, twenty-two authors, nineteen business
men, and about 130 teachers of English and of speech. 2 The
judges were asked to place the various expressions into one of
the following three categories, according to their observation
of what usage is rather than their opinion of what usage ought
to be:
1. Formally correct English, appropriate chiefly for serious and im¬
portant occasions, whether in speech or writing; usually called liter¬
ary English”
2. Fully acceptable English for informal conversation, correspond¬
ence, and all other writing of well-bred ease; not wholly appropriate
for occasions of literary dignity; “standard, cultivated colloquial Eng¬
lish”
3. Popular or illiterate speech, not used by persons who wish to pass
as cultivated, save to represent uneducated speech, or to be jocose;
here taken to include slang or argot, and dialect forms not admissible
to the standard or cultivated area; usually called “vulgar English,”
but with no implication necessarily of the current meaning of vulgar;
“naif, popular, or uncultivated English” 3
That the stipulation to “score according to your observation
of what is actual usage rather than your opinion of what usage
1 Quoted from the instructions to the judges.
2 Of these, approximately fifty were college instructors belonging to the Mod¬
em Language Association; another fifty, including many teachers from the high
schools and grammar grades, consisted of members of the National Council;
the remaining thirty were teachers of speech. There were really two lists of ques¬
tionable expressions; the first, of 102 items, was submitted to all of the judges.
This was the original list compiled by Leonard and Moffett in 1927. A second
ballot of 130 items was submitted only to the linguists and members of the
National Council.
3 There was also a fourth category, “trade or technical English” which was
employed so infrequently that it assumed no importance in the final tabulations
and is omitted from consideration here.
4 FACTS ABOUT CURRENT ENGLISH USAGE
should be” was at times more honored in the breach than in the
observance is indicated by such comments as, “I do not like very
amused”; ££ I dislike this, but rather because it is stylistically bad
than because it is grammatically incorrect”; and “One is the
proper form.” 4
PRESENTATION OF RESULTS
After these various ratings had been tabulated, the results
were presented in Current English Usage in two lists, the first,
pages 168-175, giving the ranking of all the items according to
the vote of the linguists is reproduced below. The second, pages
179-186, gave the ranking of all the items according to a per
capita vote of the whole group of judges. Each of these lists was
divided into three parts. Those items which the judges generally
agreed upon as being either literary or cultivated colloquial Eng¬
lish were labeled established . Those items which were generally
agreed upon as being uncultivated or popular English were
labeled illiterate. Finally, the expressions about which there was
marked disagreement were placed in a middle group and labeled
disputable . 5
GRAMMATICAL USAGES AS RANKED BY
LINGUISTS
A. ESTABLISHED USAGES
1. A Tale of Two Cities is an historical novel.
2. It was I that broke the vase, father.
3. Why pursue a vain hope?
4. One rarely enjoys one’s luncheon when one is tired.
5. The invalid was able partially to raise his body.
4 Current English Usage, pp. 107, 134, 153.
5 Approval as formal or cultivated colloquial English by at least 75 per cent
of the judges was required to place an item in the “established” group; disap¬
proval by at least 75 per cent of the judges was required to place an item in the
“illiterate” category. Hence “disputable” usages are those approved by more than
25 per cent but less than 75 per cent of the judges.
THE LEONARD STUDY ON USAGE 5
6. It behooves them to take action at once.
7. I had rather go at once.
8. In this connection, I should add . . .
9. This is a man . . . I used to know. (Omitted relative.)
10. You had better stop that foolishness.
11. Each person should of course bear his or her share of the ex¬
pense.
12. Galileo discovered that the earth moved.
13. This hat is not so large as mine.
14. My position in the company was satisfactory from every point
of view.
15. He toils to the end that he may amass wealth.
16. In the case of students who elect an extra subject, an additional
fee is charged.
17. The defendant’s case was hurt by this admission.
18. I for one hope he will be there.
19. This is the chapter whose contents cause most discussion.
20. Under these circumstances I will concede the point.
21. I have no prejudices, and that is the cause of my unpopularity.
22. You may ask whomsoever you please.
23. The honest person is to be applauded.
24. He stood in front of the class to speak.
25. This much is certain.
26. He did not do as well as we expected.
27. We got home at three o’clock.
28. He has no fear; nothing can confuse him.
29. There is a large works near the bridge.
30. As regards the League, let me say . . .
31. “You just had a telephone call.” “Did they leave any message?”
32. I was attacked by one of those huge police dogs.
33. The women were all dressed up.
34. This was the reason why he went home.
35. This book is valueless, that one Has more to recommend it.
(Comma splice.)
36. Take two cups of flour.
37. None of them are here.
38. I drove the car around the block.
39. He doesn’t do it the way I do.
40. The New York climate is healthiest in fall.
6
FACTS ABOUT CURRENT ENGLISH USAGE
41. I felt I could walk no further.
42. One is not fit to vote at the age of eighteen.
43. Our catch was pretty good.
44. We have made some progress along these lines.
45. The catcher stands back of the home plate.
46. My colleagues and I shall be glad to help you.
47. I went immediately into the banquet room, which was, I found
later, a technical error.
48. That will be all right, you may be sure.
49. We will try and get it.
50. We cannot discover from whence this rumor emanates.
51. I can hardly stand him.
52. Jane was home all last week.
53. I’d like to make a correction.
54. I’ve absolutely got to go.
55. We can expect the commission to at least protect our interests.
56. That’s a dangerous curve; you’d better go slow.
57. There are some nice people here.
58. Will you be at the Browns’ this evening?
59. Have you fixed the fire for the night?
60. I don’t know if I can.
61. In hopes of seeing you, I asked . . .
62. It says in the book that . . .
63. If it wasn’t for football, school life would be dull.
64. His attack on my motives made me peevish.
65. We taxied to the station to catch the train.
66. We only had one left.
67. My viewpoint on this is that we ought to make concessions.
68. Factories were mostly closed on election day.
69. He moves mighty quick on a tennis court.
70. He stopped to price some flowers.
71. He worked with much snap.
72. This room is awfully cold.
73. It is me.
74. Who are you looking for?
75. A treaty was concluded between the four powers.
76. You had to have property to vote, in the eighteenth century.
77. The kind of apples you mean are large and sour.
78. I have a heap of work to do.
THE LEONARD STUDY ON USAGE
7
79. I felt badly about bis death.
80. The real reason he failed was because he tried to do too much.
81. Invite whoever you like to the party.
82. Drive slow down that hill!
83. Harry was a little shaver about this tall.
84. I didn’t speak to my uncle by long distance; I couldn’t get
through.
85. They had numerous strikes in England.
86. I will go, providing you keep away.
87. I have got my own opinion on that.
88. He made a date for next week.
89. My father walked very slow down the street.
90. There was a bed, a dresser, and two chairs in the room.
91. They invited my friends and myself.
92. It is now plain and evident why he left.
93. I wish I was wonderful.
94. I’ve no doubt but what he will come.
95. What was the reason for Bennett making that disturbance?
96. Can I be excused from this class?
97. Haven’t you got through yet?
98. Everyone was here, but they all went home early.
99. He loaned me his skates.
100. My folks sent me a check.
101. He came around four o’clock.
102. If it had been us, we would admit it.
103. They went way around by the the orchard road.
104. The banker loaned me $200 at 6%.
105. Pikes Peak is in Colorado.
106. The sailors laid out along the yards.
107. Is your insurance sufficient coverage for your house?
B. DISPUTABLE USAGES
108. That clock must be fixed.
109. My contention has been proven many times.
110. Sam, who was then in town, was with me the three or four first
days.
111. One rarely likes to do as he is told.
112. He never works evenings or Sundays.
8
FACTS ABOUT CURRENT ENGLISH USAGE
113. They have gotten a new car this year.
114. The Rock Island depot burned down last night.
115. Sitting in back of John, he said, “Now guess what I have.”
116. I took it to be they.
117. I guess IT go to lunch.
118. He went right home and told his father.
119. He could write as well or better than I.
120. I expect he knows his subject.
121. I can y t seem to get this problem right.
122. I was pretty mad about it.
123. Either of these three roads is good.
124. You are older than me.
125. What are the chances of them being found out?
126. There is a big woods behind the house.
127. I know it to be he.
128. Do you wish for some ice cream?
129. Intoxication is when the brain is affected by certain stimulants.
130. Neither of your reasons are really valid.
131. He dove off the pier.
132. Trollope’s novels have already begun to date.
133. Will you go? Sure.
134. He is kind of silly , I think.
135. I will probably come a little late.
136. That was the reason for me leaving school.
137. They eat (et) dinner at twelve o’clock.
138. I’ll swear that was him.
139. Well, that’s going some.
140. Leave me alone, or else get out.
141. Of two disputants, the warmest is generally in the wrong.
142. It was good and cold when I came in.
143. We ha vm’t but a few left.
144. In the collision with a Packard, our car naturally got the worse
of it.
145. I wouldn’t have said that if I had thought it would have shocked
her.
146. Yourself and your guests are invited.
147. The man was very amused.
148. Such naif actions seem to me absurd.
149. It seems to be them.
THE LEONARD STUDY ON USAGE
9
150. Everybody bought their own ticket.
151. Say, do you know who that is?
152. I suppose that’s him.
153. I can't help but eat it.
154. Aren't {'nt or rnt) I right?
155. There is a row of beds with a curtain between each bed.
156. If I asked him, he would likely refuse.
157. John didn’t do so bad this time.
158. Cities and villages are being stripped of all they contain not
only, but often of their very inhabitants.
159. Everybody's else affairs are his concern.
160. It don't make any difference what you think.
161. I read in the paper where a plane was lost.
162. That boy’s mischievous behavior aggravates me.
163. That stock market collapse left me busted.
164. Neither author nor publisher are subject to censorship.
165. Yes, our plan worked just fine.
166. The fire captain with his loyal men were cheered.
167. Don’t get these kind of gloves.
168. The British look at this differently than we do.
169. Most anybody can do that.
170. It is liable to snow tonight.
171. They went in search for the missing child.
172. I suppose I’m wrong, ain't I?
173. John was raised by his aunt.
174. Martha don't sew as well as she used to.
175. He most always does what his wife tells him.
176. It sure was good to see Uncle Charles.
177. My experience on the farm helped me some, of course.
178. It’s real cold today.
179. His presence was valueless not only, but a hindrance as well.
180. We don’t often see sunsets like they have in the tropics.
181. I am older than him.
182. She leaped off of the moving car.
183. She sung very well.
184. It is only a little ways farther.
185. It looked like they meant business.
186. Do it like he tells you.
187. The child was weak, due to improper feeding.
10
FACTS ABOUT CURRENT ENGLISH USAGE
C. UNCULTIVATED OR ILLITERATE USAGES
188. John had awoken much earlier than usual.
189. I haven’t hardly any money.
190. The engine was hitting good this morning.
191. The dessert was made with whip cream.
192. Now just where are we at?
193. The kitten mews whenever it wants in .
194. A woman whom I know was my friend spoke next.
195. He drunk too much ice water.
196. Reverend Jones will preach.
197. All came except she .
198. The party who wrote that was a scholar.
199. My Uncle John, he told me a story.
200. He begun to make excuses.
201. I calculate to go soon.
202. This is all the further I can read.
203. That ain’t so.
204. The data is often inaccurate.
205. He looked at me and says . . .
206. I must go and lay down.
207. Ain’t that just like a man?
208. Both leaves of the drawbridge raise at once.
209. The people which were here have all gone.
210. I have drank all my milk.
211. That there rooster is a fighter.
212. The old poodle was to no sense agreeable.
213. One of my brothers were helping me.
214. I enjoy wandering among a library.
215. A light complected girl passed.
216. I want for you to come at once.
217. He won’t leave me come in.
218. There was a orange in the dish.
219. It was dark when he come in.
220. You was mistaken about that, John.
221. I wish he hadn’t of come.
222. Hadn’t you ought to ask your mother?
223. My cold wa’nt any better next day.
224. If John had of come, I needn’t have.
THE LEONARD STUDY ON USAGE
11
225. I had hardly laid down again when the phone rang.
226. He did noble,
227. Somebody run past just as I opened the door.
228. Just set down and rest awhile.
229. The neighbors took turns setting up with him.
230. They swang their partners in the reel.
Although it is not necessary here to consider the merits of
the practice followed in the body of the Leonard monograph,
that of giving special prominence to the rankings of the linguists,
there is no doubt that the final results were somewhat compli¬
cated by the presentation of two rankings instead of one. How¬
ever, the following chart attempts to combine in a single tabula¬
tion the two rankings given in Current English Usage , that of the
group of linguistic experts, and the per capita vote of all the
judges.
TABLE I
Distribution of Rankings in the Current English Usage Ballots
I. Expressions rated “established” by both groups. 71 items
Ila. Expressions rated “established” by the linguists and “dis¬
putable” by the vote of the whole group. 36 items
Jib. Expressions rated “'disputable” by the linguists and “estab¬
lished” by the vote of the whole group. 6 items
lie. Expressions rated “disputable” by both groups. 64 items
lid. Expressions rated “disputable” by the linguists and “illit¬
erate” by the vote of the whole group. 10 items
He, Expressions rated “illiterate” by the linguists and “disputa¬
ble” by the vote of the whole group. 5 items
III. Expressions rated “illiterate” by both groups. 38 items
Total 230 items
To sum up the matter very briefly, seventy-one of the 230 test
expressions were rated as acceptable or established both by the
vote of the linguists considered separately and by the per capita
vote of all the judges. Likewise these two groups also agreed in
’ condemning as illiterate thirty-eight of the total number. This
12 FACTS ABOUT CURRENT ENGLISH USAGE
accounts for 109 of the 230 expressions. The status of the remain¬
ing 121 expressions is left somewhat in doubt since either one
group of judges or both groups were unable to come to any agree¬
ment about them.
II e£L
THE PURPOSE AND METHOD OF THIS STUDY
At this point in our analysis of Current English Usage we come
to the purpose of the present study. It has been pointed out that
Leonard and his associates made a survey of opinion about usage
rather than the facts of usage. This appeal to opinion may be
said to have had decisive results in less than one-half of the cases
submitted to the judges. Both of these considerations suggest an
avenue for further study of the same material: a compilation of
the recorded facts of usage concerning the same 230 expressions
and a comparison of these facts with the collected opinions repre¬
sented by the Leonard report.
The attitude of the compilers of the Leonard report as to the
reliability of available collections of linguistic fact is not wholly
clear. Although the Oxford Dictionary and other dictionaries are
frequently quoted in the detailed treatment of the various items,
one finds in the introduction to the grammar section the follow¬
ing statement, justifying a survey of opinion rather than of fact:
Dictionaries have as their prime function the recording of usage,
but by their very nature most of their citations have to be drawn
from literary examples of acknowledged value; this method, valid
though it may be, must of necessity result in a lag of several years
between the adoption of a given usage and its appearance in a diction¬
ary. 6
On the other hand, the following excerpts from Leonard’s
original article indicate clearly enough that he intended his sur-
6 Current English Usage, p. 95.
13
14
FACTS ABOUT CURRENT ENGLISH USAGE
vey of opinion to be ancillary to, rather than a substitute for, a
survey of fact.
1. The net result of the study is to make possible some interesting
comparisons between (1) current statements by grammar and rheto¬
ric texts, and particularly by the school of grammarians and rheto¬
ricians which rose in the eighteenth century and has continued to
this day, and (2) the record of actual cultivated usage as shown by
this study, and by the complete dictionaries, particularly the Oxford.
2. It is submitted, however, that this record and that of the
dictionary should help to set purists and their purism in a correct
light.
3. In particular such data should be taken into account in decid¬
ing between the records of the dictionary-makers and those of the
writers of textbooks whenever the latter purport to state what is
actually used. 7
It is, then, wholly in keeping with Leonard’s conception of his
survey to compare his findings, based on opinion, with available
compilations of recorded usage, that is to say, recorded fact. The
detailed treatments of the separate items in the body of Current
English Usage do make such comparisons upon a great many oc¬
casions, but at the time this monograph was written neither the
second edition of Webster’s New International Dictionary nor
the Supplement to the Oxford Dictionary was available. More¬
over, information about a considerable number of the problems
involved in the test expressions is not of a nature which lends
Itself to convenient or systematic presentation in a dictionary,
but is, nevertheless, contained in other collections of linguistic
fact.
It is particularly important to secure factual information
about the status of the “disputable” expressions, those concern¬
ing which the survey of opinion gave an indecisive answer. It
will be recalled that the judges—or the jury, properly speaking
—disagreed in 121 cases out of 230. A jury disagreement is, of
7 S. A. Leonard and H. Y. Moffett, “Current Definition of Levels ip English
Usage,” English Journal^ Vo! 16, May, 1927, pp, 356-35?.
THE METHOD OF THIS STUDY
IS
course, seldom a satisfaction or a help to any one, least of all to
the teacher who is attempting to fashion his classroom practice
on the basis of its findings. True enough, the Leonard study sug¬
gests in its conclusion that the teacher “will certainly, in mark¬
ing themes, accept from the average student any usage classed in
this study as established or disputable,” 8 but in the very next
sentence it is hinted that certain other usages may be more
elegant than those labeled disputable . At any rate, the word
disputable has a somewhat dubious sound and has no appropriate
place in the description of usage, which is linguistic fact. A usage
may be established; it may be popular; it may be regional; it
may be upper or lower class; but strictly speaking, it cannot be
disputable. Only opinion about it may be suitably described by
such a term. When opinion fails to give a satisfactorily definite
answer concerning the status of one or more expressions, the only
thing that remains to be done, or that can be done, is to look at
the facts.
The purpose of the present study, then, is to supplement the
survey of opinion, which forms the basis of the Leonard mono¬
graph, with a survey of the recorded usage of the same 230 items.
It has already been pointed out (1) that Leonard had no inten¬
tion of excluding from consideration such a survey of fact; (2)
that new sources of factual information have become available
since the Leonard report was compiled; and (3) that in more
than half the cases considered, the survey of opinion failed to
give a decisive answer as to the status of the expressions. These
considerations taken together seem to the authors to justify the
present supplementary investigation.
THE METHOD IN THIS STUDY
Having outlined the purpose of the present study, one is im¬
mediately confronted with the problem of sources: What cqJ-
$ Current Engljsji ZJ$pge } p. 1§8.
16 FACTS ABOUT CURRENT ENGLISH USAGE
lections of linguistic fact are to be employed In carrying it out?
It will be accepted without argument; we believe; that the most
convenient and authoritative single compilation of linguistic fact
Is the Oxford Dictionary together with its Supplement. Accord¬
ingly; the Oxford Dictionary was consulted In respect to each of
the test expressions; 9 to discover what record there was of Its
use on the formal literary level; on the informal or colloquial
level, in dialect, at the present time, or in any earlier period. In
a great many instances the Information given In the Oxford
Dictionary was deemed sufficiently complete for the purposes
of this investigation. Since this monumental work was a long
time in the making, some of the earlier volumes are based on
less complete evidence than those in the second half of the
alphabet. Thus, It was felt desirable, at times, to add to the Ox¬
ford Dictionary data, information supplied by the second
(1934) edition of Webster’s New International Dictionary .
This was especially true in the case of words or expressions
whose status In usage is not the same in Great Britain and
America. Horwill’s Modern American Usage 10 was also con¬
sulted in respect to suspected Americanisms.
On a number of occasions the test expressions in the Leonard
study were included for the purpose of presenting syntactical
issues, as for example No. 141 In the linguists’ ranking, which
was designed to raise the question of the use of the superlative
degree of an adjective in place of the comparative. 11 The es¬
sential point here is not the extent of the use of any single adjec¬
tive in this fashion, but rather, usage in general in respect to
this particular function of the adjective. In such instances the
dictionary was not the most satisfactory source for a record of
9 This statement stands in some need of qualification. In the portion of the
investigation conducted by Walcott, namely the status of the “established” ex¬
pressions, it was felt that to gather factual data concerning such unquestionably
accepted expressions as “It is I,” would be a work of supererogation.
10 h. W. Horwill, Modern American Usage (Oxford, Clarendon Press, 1935).
11 “Of two disputants, the warmest is generally in the wrong.” Current Eng¬
lish Usage, p. 139,
THE METHOD OF THIS STUDY
17
usage, and the grammars of Jespersen and Curme were used
to supplement the dictionary findings. 12 These grammars have
the weight of scholarly authority behind them, and in them
illustrative citations from modern writings and from earlier
periods are extensively employed to support the observations
which are made. Likewise Hall’s English Usage, 13 a record of
usage based upon seventy-five thousand or more pages of literary
English, was consulted at various points. In a few other in¬
stances articles in the scholarly journals were employed.
RESULTS OF THE INVESTIGATION
The results of this investigation may be most conveniently
presented in condensed form, and a few words will suffice to
explain the subsequent arrangement and treatment of the ma¬
terial. First of all, the 230 test expressions were divided into the
three groups indicated in Table I (“established,” “disputable,”
or “illiterate”) on the basis of their combined rating in the two
ballots. The authorities named previously were consulted in re¬
spect to each of these expressions. When the expression was
found recorded in one or more of the collections of usages em¬
ployed, it was placed in one of six categories, Literary English,
American Literary English, Colloquial English, American Col¬
loquial English, Dialect, or Archaic. The data concerning the ex¬
pressions in each of the three groups were then combined. Since
the record of usage in each of the groups, when compared with
the Leonard survey of opinion, raises a number of questions and
presents certain problems somewhat peculiar to itself, the three
groups are considered separately in this study.
A few explanatory comments about the method of classifica¬
tion are in order here also. If the expression was recorded with-
12 O. Jespersen, A Modern English Grammar (Heidelberg, 1928-31), 4 vols.
G. 0. Curme, Syntax (Boston, D. C. Heath and Co., 1931).
-, Parts of Speech and Accidence (Boston, D. C. Heath and Co., 1935).
13 J. Leslie Hall, English Usage (Chicago, Scott, Foresman and Co., 1917).
18
FACTS ABOUT CURRENT ENGLISH USAGE
out a limiting label in the collections of usage consulted, and
if there was at least one citation from the nineteenth century,
the expression was considered Literary English . The Oxford
Dictionary , extensive as it is, found it necessary to limit the
number of citations to one for each century—although in some
cases more were used—and all of the volumes through the
letter / had appeared by 1900. On the other hand, most of the
nineteenth-century citations found were well within the second
half of the century, so that there is little danger of an archaism
being recorded here as in present use.
It must be admitted, of course, that if the last Oxford Dic¬
tionary citation for a certain expression happens to be 1875, a
critic of this study might point out that usage could have
changed in the last sixty years, that what was acceptable Eng¬
lish in 1875 might have become unpalatable in the second quar¬
ter of the twentieth century. This is entirely possible, yet one or
two considerations may be brought forward in defence of the
procedure adopted here. First, if we may anticipate our results
somewhat, the great majority of the supposed errors included
in the test items were discovered, upon examination, to be
usages which had been in the language for centuries, a number
of them going back to the very beginnings of recorded English.
To take a specific instance, the use of between referring to more
than two (for example, A treaty was concluded between the
four powers) is recorded in literary usage as early as 971, while
the last citation given is 1885. While it is possible that the last
fifty years might have seen a change in the status of this use of
between , yet its continuous history of nine hundred years 5 stand¬
ing would seem to militate against this supposition. It should be
remembered also that the Oxford Dictionary editors had many
more citations for most words than they could possibly use,
and that citations were selected for inclusion upon considera¬
tions other than date. Thus if the last citation for a word bears
the date 1875, this does not necessarily indicate that it is the
THE METHOD OF THIS STUDY
19
most recent one that the editor had in his possession; it is merely
the one he chose to have printed.
To return to the bases of classification, if an expression was
given in any of the sources with the limiting label U. S . and the
citations illustrating it were drawn from serious or formal writ¬
ing, it was classified as American Literary English. The cate¬
gories Colloquial English and American Colloquial English are
self-explanatory. It must be emphasized, however, that the term
colloquial, as it is employed in reputable dictionaries and by
sound scholars, is not used in a derogatory sense. It merely
means that the expression or word is to be found in spoken or
informal written rather than in formal written English.
This division of items into Literary English and American
Literary, Colloquial English and American Colloquial is in¬
complete in one respect. There is no provision for those usages
which are peculiarly British English, since it should be assumed
that the unqualified terms Literary English and Colloquial Eng¬
lish imply acceptability on both sides of the Atlantic. However,
since only three of the 230 items turned out to be Briticisms, the
authors felt that a separate category for these would scarcely
be necessary.
The Dialect category needs but one bit of explanation. Since
the Oxford Dictionary recorded only dialectical words or ex¬
pressions that had formerly been in general use, not those that
had begun as dialect and remained so, the same qualification
must apply to what is listed as dialect here. Any words or ex¬
pressions for which no citations after 1800 were found in the
Oxford Dictionary were listed as Archaic unless one of the
other sources indicated that they were still in present use. If a
word was labeled both dialect and archaic by the dictionaries, it
was placed in the dialect category on the ground that it was still
in use somewhere at the present time.
In contrast to the Leonard survey, which is based wholly
and frankly upon subjective impression, the present authors
20 FACTS ABOUT CURRENT ENGLISH USAGE
have attempted to make their analysis as objective as possible.
In general, the facts have been allowed to speak for themselves.
The subjective element enters in only when the item under con¬
sideration is placed in one of the various categories, Literary
English, Dialect, etc., and here it was occasionally necessary
to differ with the conclusions reached by the dictionary editors
and grammarians. If an expression was recorded without com¬
ment, but the citations appeared to have been drawn from in¬
formally written material or from dialogue, it was recorded as
Colloquial. If, on the other hand, an expression was labeled Col¬
loquial but the citations were chiefly drawn from works of a
serious literary nature, the authors felt justified in classifying
such an item as Literary English. In the case of disagreement
among the various authorities, the only feasible solution seemed
to be to classify the expression on the basis of the citations that
were given.
In a few instances, notably those of like as a conjunction and
whoever used obliquely, the dictionaries commented that al¬
though the expression was condemned by some, it was neverthe¬
less to be found in many recent writers of standing. Because
of this very definite statement concerning extent of usage, the
only consistent policy in such cases seemed to be to assign
the expressions to the literary category. The authors felt that
the reported facts of usage deserved greater consideration than
the reported attitude toward the expression. Likewise dictionary
judgments as to grammatical correctness were considered of
less weight than statements concerning extent of use; the former
are frequently omitted from our direct quotations from the
dictionaries.
Before proceeding to the detailed analysis of our findings
one more thing remains to be said. In some respects the results
of this study differ from certain widely held notions about
present-day English to an even greater extent than the Leonard
report. Therefore, it is possible that our work will meet with
THE METHOD OF THIS STUDY 21
much of the same criticism that greeted the appearance of Cur¬
rent English Usage. The authors wish to make it clear at this
point that in this study they are not advocating any one usage,
a group of usages, or a level of usage. With advocacy of any
kind we have absolutely no concern. We are only reporting the
facts of the English language as they appear in the work of
universally recognized authorities. We have conscientiously
given all of the sources of our information so that our findings
may be verified. The subjective element has been eliminated as
far as was humanly possible, that is except for placing the items
into their respective categories, and in performing this task we
have attempted to lean neither to right nor left.
Moreover, it should be understood that the question of usage
as a proper basis for grammar does not at all enter into this
study, which concerns itself only with collecting the record of
usage of a given number of items. Incidentally, these items were
not chosen by the present authors. Nor are we attempting to
fix as standard usage or to disguise as current English the oc¬
casional errors of one or a number of authors. In the first place,
the term error implies a judgment based upon an already ac¬
cepted standard, something quite outside the scope of our study.
In the second place, the responsibility of recording the oc¬
casional error or supposed slip rests with the authorities we
employed, not with us. Our only concern is to compare the
evidence of language usage as it appears in the monuments of
scholarship esteemed by all of us with the impressions of Leon¬
ard’s judges, and to see what may be learned from such a com¬
parison.
III
THE “ESTABLISHED” USAGES
The seventy-one items in the Leonard monograph which re¬
ceived a vote of “established” both in the ballot of the linguists
and the per capita vote of the whole group are as follows: Nos.
1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, IS, 16, 17, 18, 19, 20,
21, 22, 23, 24, 25, 26, 28, 31, 32, 33, 34, 36, 38, 39, 40, 41, 42,
43, 44, 45, 46, 47, 48, 51, 52, S3, 54, 55, 56, 57, 58, 59, 60, 63,
64, 65, 70, 71, 72, 76, 77, 80, 83, 84, 85, 99, 103, 104, 105, 106,
107. 14
Of the seventy-one established items, nine may be either
classed as hyper-urbanisms, or else they are such as would
seem to conform to the traditional standard of grammatical cor¬
rectness to the extent of being almost non-controversial. That
these nine expressions should be intermingled with the other
items might be considered somewhat ill advised anyway, for
any judgments concerning them entail for the most part a re¬
versal of the critical issue, from proscribing what might appear
as questionable to disapproving what might seem too fastidious.
The nine examples follow in their numerical order:
1. A Tale of Two Cities is an historical novel.
2. It was I that broke the vase, father.
4. One rarely enjoys one’s luncheon when one is tired.
5. The invalid was able partially to raise his body.
6. It behooves them to take action at once.
14 The numbering here is based on the ranking given the item in the vote of
the linguists; the same scheme was used in the body of Current English Usage.
22
THE “ESTABLISHED” USAGES 23
11. Each person should of course bear his or her share of the ex¬
pense.
13. This hat is not so large as mine.
22. You may ask whomsoever you please.
46. My colleagues and I shall be glad to help you.
It is perhaps sufficient to note that 72 per cent of all the
judgments upon these nine items fall within the literary cate¬
gory; 23 per cent class them as colloquial; and only 5 per cent
brand them as illiterate. 15 Surely these are not questions of
active controversy, nor are they of immediate concern to
teachers of English expression.
A similar group may also be tentatively distinguished from
the class’of “usually condemned” usages; they represent shades
of distinction entirely too subtle for any but the more fastidious
of literary stylists to differentiate. The list follows:
3. Why pursue a vain hope?
8. In this connection, I should add . . ,
14. My position in the company was satisfactory from every point
of view.
15. He toils to the end that he may amass wealth.
16. In the case of students who elect an extra subject, an addi¬
tional fee is charged.
17. The defendant’s case was hurt by this admission.
18. I for one hope he will be there.
20. Under these circumstances I will concede the point.
21. I have no prejudices, and that is the cause of my unpopularity.
23. The honest person is to be applauded.
24. He stood in front of the class to speak.
25. This much is certain.
28. He has no fear; nothing can confuse him.
15 These percentages are derived by adding together the total number of opin¬
ions 4 upon this entire group of items that fall within each of the three respective
categories: literary, colloquial, and illiterate English; and then dividing each of
the three totals by the entire number of judgments passed upon the composite
group. The figures are not offered with any mistaken notion of precise mathe¬
matical proportions. They merely reflect a fairly accurate summary of the gen¬
eral attitudes of the judges toward the composite group which we have chosen
to consider separately.
24
FACTS ABOUT CURRENT ENGLISH USAGE
32.1 was attacked by one of those huge police dogs.
44. We have made some progress along these lines .
45. The catcher stands back of the home plate.
47. I went immediately into the banquet room, which was, I found
later, a technical error.
48. That will be all right , you may be sure.
64. His attack on my motives made me peevish .
Here we have nineteen items that can scarcely be said to
involve grammatical problems at all. The printed comments
upon them are significant of this. The collective tabulations
in the case of this group are again significant. The judges divide,
for the most part, between placing them in the accepted literary
category (38.4 per cent) or in the cultivated colloquial group
(54.9 per cent). Only 6.7 per cent of all the judges’ opinions
upon these respective items would place them definitely within
the class of illiterate expression. The ultra-stylistic nature of
these problems, together with the significant expression of the
judges upon them, should surely establish the group as relatively
unimportant for classroom emphasis.
There is still another class of usage that is admittedly tech¬
nical, usage that has become current in some specialized de¬
partment of affairs and that the literary writer must almost of
necessity adopt whenever he would invade the new field of
denotations. The editors place four items of the seventy-one
within this special class. 16
104. The banker loaned me $200 at 6 per cent.
105. Pikes Peak is in Colorado.
106. The sailors laid out along the yards.
107. Is your insurance sufficient coverage for your house?
16 It it interesting to note that of these four items the judges recognized only
the last as definitely established in technical use. The editors, after remarking the
general misunderstanding occasioned by Nos. 104 and 106, arbitrarily included
these two within the established group on the authority of the dictionaries, and
contrary, in the case of the latter item at least, to the overwhelming opposition
of the judges.
THE “ESTABLISHED” USAGES 25
There are also some other expressions that might be called
colloquially technical —that is, they have come to represent
meanings in colloquial use for which satisfactory literary sub¬
stitutions are superfluous, if not impossible. The following
usages may be classified in this fashion:
36. Take two cups of flour.
38. I drove the car around the block.
65. We taxied to the station to catch the train.
84. I didn’t speak to my uncle by long distance; I couldn’t get
through.
The judges are very tolerant of these four usages. When they
are taken collectively, 12.7 per cent of all the opinions are for
inclusion within the literary category, as against 6.9 per cent
for the illiterate. The greater proportion of the judgments, 80.4
per cent, place them within the cultivated colloquial group.
We have now reduced the list of seventy-one “established”
usages to thirty-five items about which a considerable portion
of our more active language controversies are likely to arise.
The major aim of the present section of this study will be to
consider this group of thirty-five language problems from the
point of view of actual evidence of usage, either literary or
colloquial, in order to substantiate or to refute the findings of
the questionnaire by material proof.
The evidence in respect to the use of each of these items is
presented here in condensed form. In each case the expression
is quoted in full with the point in question italicized; the rank¬
ing in the two Current English Usage ballots is given, the dates
of the earliest and latest citations which have been found, and
any additional information concerning the spheres or limitations
of use is also given. It should be understood, also, that when
certain authors are cited as using a particular locution, the
evidence is not at all confined to the writers mentioned. We have
merely selected representative names to give the reader a gen-
26
FACTS ABOUT CURRENT ENGLISH USAGE
eral idea of the range and extent of use. The letters at the
extreme right margin following the resume of evidence indicate
the category in which the authors have placed the expression
in question, as Literary English, Colloquial English, etc. It may
be pointed out again that the application of such labels repre¬
sents ultimately a subjective judgment, but that at all times
the authors have conscientiously attempted to make this judg¬
ment depend on the evidence before them.
To facilitate reference to the evidence in the Oxford Dic¬
tionary , the number of the definition under which the particular
use of the word or expression is treated has been given, immedi¬
ately preceding the dates of the citations. In the case of forms
of verbs, this practice could not be followed since such forms
were frequently scattered throughout the treatment of the word.
Key to Abbreviations
L followed by a number indicates the ranking in the bal¬
lot of the linguists (see pages 4-12).
WG followed by a number indicates the ranking in the
ballot of the whole group of judges.
OD —Oxford Dictionary
OD Sup —Oxford Dictionary Supplement
Jesp —Jespersen, Modern English Grammar (quoted by
volume and page)
Curme PSA —Curme, Parts of Speech and Accidence
Hall EU —Hall, English Usage
Pooley —Pooley, Grammar and Usage in Textbooks on
English
LE —Literary English
ALE —American Literary English
CE —Colloquial English
ACE —American Colloquial English
Dial —Dialect
Arch —Archaic
THE “ESTABLISHED” USAGES
27
ESTABLISHED USAGES
L 7 : WG 30 . I had rather go at once. LE
OD, s.v. rather adj. 9 d, 1450-1875.
Jesp III, 35, cites Defoe, Thackeray, Shaw, Wells.
Hall EU, 116-121, names 32 authors who use it.
L 9 : WG 19 . This is a man ... I used to know. LE
(Omitted relative.)
Jesp III, 133, cites examples from 906
to present time.
L 10 : WG 28 . You had better stop that foolishness. LE
OD, s.v. better 4 b, 971-1875, including
Shakespeare.
L 12 : WG 29 . Galileo discovered that the earth
moved. LE
Jesp IV, Chap. XI, cites Chaucer, Shake¬
speare, Bunyan, Defoe, Franklin, Goldsmith, Bos¬
well, Hunt, Dickens, Shaw, Bennett, Wilde, Gals¬
worthy, and Barrie.
L 19 : WG 22 . This is the chapter whose contents
cause most discussion. LE
OD, 3, 1382-1896, including Shakespeare
and Milton. Jesp III, 129, cites Marlowe, Shake¬
speare, Shelley, Stevenson, and others. Hall EU,
320-327, lists 150 authors, from Malory to present
day, who use the expression.
L 26 : WG 33 . He did not do as well as we expected. LE
Storm, Englische Philologie, 696, cites
Swift, Johnson, Boswell, Burney, Dickens, Marryat,
Trollope, and others.
L 31 : WG 58 . “You just had a telephone call.”
“Did they leave any message?” LE
OD, 2 , 3, 1415-1896. Jesp III, 137, “fre¬
quent.” Cites Thackeray.
28
FACTS ABOUT CURRENT ENGLISH USAGE
L 33 : WG 45 . The women were all dressed up. LE
OD, s.v. dress vb. 7 d, 1674-1721. Jesp
III , 330. Webster records without comment.
L 34 : WG 49 . This was the reason why he went
home. LE
OD, s.v. why S a, 1225-1908.
L 39 : WG 31 . He doesn’t do it the way I do.
OD, 14, 725-1897, cites Shakespeare,
Addison, Hardy, and others.
L 40 : WG 71 . The New York climate is health¬
iest in fall.
OD, s.v. healthy 2 , 1552-1871.
L 41 : WG 50 . I felt I could walk no further. LE
OD, adv. 1,1000-1855. Curme, Syntax.
502: “In adverbial function farther and further are
used indiscriminately.” See also Pooley, 128-130.
L 42 : WG 36 . One is not £t to vote at the age of
eighteen. LE
OD, adj. 4, 1573-1868, Shakespeare,
Franklin, Browning, and others cited.
L 43 : WG 34 . Our catch was pretty good. LE
OD, adv. 1, 1565-1896. Hall EU, 217-
219, cites 52 authors.
L 51 : WG 54 . I can hardly stand him. LE
OD, 59 a and b, 1626-1891, including
Steele, Chesterfield, Carlyle. Two examples from
Mrs. Oliphant used with reference to persons; the
others are not.
L 52 : WG 46 . Jane was home all last week. CE
OD, adv. 2 , 1587-1885, although it is
doubtful whether any of the examples clearly obviate
the sense of motion toward or result of motion toward.
Curme, Syntax, 145, “popular speech.”
THE “ESTABLISHED” USAGES 29
L 53 : WG 40 . I’d like to make a correction. LE
Jesp IV, 315, cites Thackeray, Kipling,
Wilde, and D. H. Lawrence. Curme, Syntax, 368,
cites Mrs. H. Ward.
L 54 : WG 72 . I’ve absolutely got to go. LE
OD, s.v. get, 24, 1876-1889, Ruskin
cited. Jesp IV, 51, cites Disraeli, Dickens, Eliot,
Ruskin, Wilde, Shaw, Wells, Trollope. Hall EU,
121-123, cites 13 authors. OD labels this “colloq.
or vulgar,” but in view of the authors cited it is
classified as literary here.
L 55 : WG 48 . We can expect the commission to
at least protect our interests. LE
Curme, Syntax, 458 ff., cites innumer¬
able examples from the fourteenth century to the
present. Hall EU, 266-275, cites 34 authors who
use it.
L 56: WG 52 . That’s a dangerous curve; you’d
- better go slow . LE
OD, adv. 1, 1500-1858, including
Shakespeare, Milton, Byron, and Thackeray.
L 57 : WG 41 . There are some nice people here. LE
OD, adj. 15, 1769-1897.
L 58 : WG 42 . Will you be at the Browns’ this
evening? LE
Jesp IV, 255, “now increasingly fre¬
quent.” C. C. Fries, “The Periphrastic Future with
Shall and Will in Modern English,” Publications
of the Modern Language Association, XL (1925),
963 ff., finds in English drama from 1557 to 1915
a total of 505 cases of will to 7 of shall in second-
person questions.
L 59 : WG 47 . Have you £xed the fire for the
night?
OD, 14 b, 1769-1891, including Mar-
ACE
30
FACTS ABOUT CURRENT ENGLISH USAGE
ryatj Holmes, comments “chiefly U.S. colloq.” Jesp
111, 244, “in US = Repair . 5 ”
L 60 : WG 65 . I don’t know 1/ I can. LE
OD, conj. 9, Beowulf-1895, including
Caxton, the Bible , and Dryden. Jesp Ill, 42, cites
Shakespeare, Byron, and present-day authors.
L 63 : WG 68 . If it wasn’t for football, school life
would be dull. LE
Jesp IV, 130, cites Marlowe, Spectator,
Sheridan, Austen, C. Bronte in constructions with
the negative. C. Alphonso Smith, “The Indicative
in Unreal Condition,” Modern Philology, V, 361,
gives earliest examples from fifteenth-century Bat¬
tle of Otterburn and follows with others from Pepys,
Bunyan, Defoe, Burke, Sheridan, a host of nine¬
teenth-century writers, and comments, “By no
means confined to colloquial language.”
L 70 : WG 51 . He stopped to price some flowers. LE
OD, vb. 3, 1845-1872.
L 71 : WG 69 . He worked with much snap. LE
OD, sb. 11, 1872-1894.
L 72 : WG 70 . This room is awfully cold. CE
OD, 3, 1816-1878, comments “slang,”
but examples are apparently colloquial rather than
slang.
L 76 : WG 57 . You had to have property to vote,
in the eighteenth century. LE
OD, 6 , 1577-1870. Bacon, Swift, and
Ruskin cited.
L 77 : WG 61 . The kind of apples you mean are
large and sour. CE
OD, s.v. kind, sb. 14 b, 1382-1797, cit¬
ing Wycliff and Shakespeare, comments “still com¬
mon colloquially.” Jesp II, 66, and Curme, Syntax,
544, label this construction colloquial.
THE “ESTABLISHED” USAGES
31
L 80 : WG 66 . The real reason he failed was be¬
cause he tried to do too much. LE
OD, s.v. because, conj. 2 , 1656, com¬
ments “Obs” (common dial.). Pooley, 119, cites
the construction in speeches by Stuart Chase and
Curme, and in writings by Edith Franklin Wyatt,
Clarence Day, and Frank Harris. E. E. Ericson,
Anglia, 41 (1937), 112-113, cites Thomas Hobbes,
W. L. Phelps, W. P. Trent, John Macy, the New
York Times, and Baltimore Sun.
L 83 : WG 14 . Harry was a little shaver about
this tall. LE
OD, adv. 2 b, 1460-1885.
L 85 : WG 67 . They had numerous strikes in
England in i 860 . CE
OD, 3, 1415-1896, Shakespeare and
Jonson cited. “Much used colloquially and dialecti¬
cally.”
L 99 : WG 63 . He loaned me his skates. ALE
OD, vb., 1200-1901, but comments,
“Now chiefly U.S” Hall EU, 150-151. Horwill,
Modern American Usage, 192, “In America it is
still a verb.” Webster records without comment.
L 103 : WG 73 . They went way around by the or¬
chard road. ACE
OD, adv. 2 , 1849-1908, “obs. exc. Sc.,
north., and U.S.” Webster, “dial, or colloq.”
The detailed evidence which has been presented for the
thirty-five expressions in the “established” group which might
be considered at all controversial may be summarized in the
following manner:
1. Twenty-nine of the thirty-five moot usages appear to satisfy
most of the demands of formal literary English. These are as fol¬
lows: Nos. 7, 9, 10, 12 , 19, 26, 31, 33, 34, 39, 40, 41, 42, 43, 51,
53, 54, 55, 56, 57, 58, 60, 63, 70, 71, 76, 80, 83, 99.
32 FACTS ABOUT CURRENT ENGLISH USAGE
2. Four of the usages, Nos. 52, 72, 77, and 85 are definitely re¬
stricted to colloquial and informal written English.
3. Two items, Nos. 59 and 103, appear to be in acceptable col¬
loquial use in America but not in England.
The most significant aspect of these conclusions lies in the
relatively few colloquialisms to be found in the “established”
group. In compiling the results of the judges’ ballots, Leonard
and his associates counted as “established” those items which
75 per cent or more of the judges had marked as appropriate
to formal literary or colloquial English. A survey of actual
usage, however, shows that even of the thirty-five most dubious
of the “established” items, only six are restricted to the col¬
loquial or informal level. Thus we have our first inkling of the
conservatism of opinion when compared with the recorded facts
of usage.
IV ed
THE “DISPUTABLE” USAGES
It has already been remarked that those usages which the
editors of the Leonard study labeled disputable are simply those
about which the judges were unable to agree. To repeat, the
appellative disputable is not appropriate in the description of
a linguistic fact; it merely is an indication of the extreme varia¬
tion of opinion. To illustrate, item No. Ill in the list, “One
rarely likes to do as he is told/ 5 was considered appropriate for
formal literary usage by six linguists but illiterate by ten; on
the other hand, thirty members of the Modern Language As¬
sociation approved it for formal literary usage, whereas only
two condemned it as illiterate. Along with several other items it
ranked second in the vote of the English Council but thirty-
third in the vote of the linguists. Obviously when opinions about
acceptability vary to such an extent, it is not only important but
absolutely essential to have recourse to the facts.
The evidence of usage as to the 121 disputable items is pre¬
sented below; these items are considered in groups as indicated
by the subdivisions of class II in Table I.
DISPUTABLE USAGES
A. LINGUISTS, “ESTABLISHED”; WHOLE GROUP “DISPUTABLE.”
L 27 : WG 79 . We got home at three o’clock. CE
OD, s.v. get, vb. 25, 1300-1712; s.v.
home, adv. 7, 1806-1886. The citations appear to
33
34 FACTS ABOUT CURRENT ENGLISH USAGE
be informal rather than literary even though they
are recorded without comment.
L 29 : WG 89 . There is a large works near the
bridge. LE
OD, s.v. work, sb. 18, 1882-1898 (with
the indefinite article), Webster, “often construed as
a singular.’’ Jesp 11, 152, quotes Masefield ( 2 ) and
Bennett.
L 30 : WG 81 . As regards the League, let me
say. . . LE
OD, s.v. regard, vb. 7 c, 1824-1885.
Jesp 111, 178, characterizes this as a “frequent com¬
bination.”
L 35 : WG 89 . This book is valueless, that one
has more to recommend it. (Comma
splice.) LE
Naturally a problem such as the comma
splice, which is the point of this expression, would
not be treated in the sources employed. Summey,
Modern Punctuation, 79, asserts that this punctua¬
tion “is manifestly growing in favor.”
L 37 : WG 84 . None of them are here. LE
OD, 2 b, 888-1887. The earliest in¬
stance with a plural verb is not 1580 as asserted in
Current English Usage, 104.
L 49 : WG 99 . We will try and get it. LE
OD, s.v. try, vb. 16 b, 1686-1883. This
expression is labeled colloquial in the OD but the
citations are drawn from reputable literary sources,
including Milton and Coleridge. It is not so labeled
in Webster where Milton’s use of the expression
is cited. Hall EU, 309, cites twenty-eight instances
from ten authors.
L 50 : WG 103 . We cannot discover from whence
this rumor emanates. LE
OD, s.v. whence, 1377-1887.
THE “DISPUTABLE” USAGES
35
L 61 : WG ioo. In hopes of seeing you, I
asked .... LE
OD, s.v. hope, sb. 1 c, 1659-1702 (in
phrase given here). Plural of the substantive cited
as late as 1864. Recorded in Webster as present
usage.
L 62 : WG 101 . It says in the book that .... CE
OD, 3 f, 1175-1482. Labeled “modern
colloquial.” OD Sup has examples up to 1927.
L 66 : WG 96 . We only had one left. LE
OD, adv. 1 c. 1483-1875, including
Dryden and Tennyson, with comment, “Frequent
in speech,” but citations are definitely literary in
character. Hall EU, 187-193, gives a list of 400 in¬
stances from 104 authors to illustrate this construc¬
tion.
L 67 : WG 85 . My viewpoint on this is that we
ought to make concessions. LE
OD, a, 1856-1892.
L 68 : WG 82 . Factories were mostly closed on
election day. LE
OD, citations with mostly in a position
comparable to this, 1594-1719; mostly with the
meaning in question is cited as late as 1904.
L 69 : WG 93 . He moves mighty quick on a ten¬
nis court. LE
OD, adv. 1 , 1290-1874, Shakespeare,
Milton, Tennyson, and others cited.
L 73 : WG 124 . It is me. CE
OD, pron. 6 , 1591-1758, Shakespeare
and Goldsmith cited. Webster, “colloquial and dia¬
lect.”
L 74 : WG 131 . Who are you looking for? CE
OD, pron. 5, 1450-1881, “common in
colloquial use.” Shakespeare, Southey, and Hardy
cited.
36
FACTS ABOUT CURRENT ENGLISH USAGE
L 75 : WG 112 . A treaty was concluded between
the four powers. LE
OD, s.v. between, prep. 19, 971-1885.
L 78 : WG 104 . I have a heap of work to do. CE
OD, sb. 4, 1661-1884, “Colloq. 75
L 79 : WG 88 . I felt badly about his death.
OD, 7, 1783-1821, “Dial. 57 Webster,
“Dial. 75
L 81 : WG hi. Invite whoever you like to the
party. LE
OD, 3, 1592-1780, including two quota¬
tions from Shakespeare. Webster, “common collo¬
quially and still found in good writers”
L 82 : WG 97 . Drive slow down that hill! LE
OD, adv. 1 , 1500-1858, Shakespeare,
Milton, Byron, and Thackeray cited.
L 86 : WG 105 . I will go, providing you keep away. LE
OD, b, 1632-1874.
L 87 : WG 114 . I have got my own opinion on that. LE
OD, s.v. get, vb. 24, 1596-1878, “in
familiar language . 77 Shakespeare, Johnson, and
Thackeray cited. Jesp IV, 48, “the 19th c. examples
show its extension to higher forms of literature , 77
citing among others Scott, Austen, Thackeray,
Morris, Ruskin, Wilde, and Shaw. Webster records
and cites Herbert.
L 88 : WG 115 . He made a date for next week. ACE
OD Sup, sb . 2 2 c, 1896-1928, “U.S.
colloq . 77
L 89 : WG 83 . My father walked very slow down
the street.
See L 82 above.
LE
THE “DISPUTABLE” USAGES
37
L 90 : WG 121 . There was a bed, a dresser, and
two chairs in the room. LE
Jesp II, 181-182, 1470-1909. Hall EU,
53-58, gives examples from Beowulf, Malory, Shake¬
speare, Bible, Milton, Thackeray, and others.
L 91 : WG 109 . They invited my friends and my¬
self. LE
OD, 3, 1205-1856, “in an enumeration,
when not occupying first place . . . commonly pre¬
ferred to me.”
L 92 : WG 106 . It is now plain and evident why he
left. CE
OD, s.v. plain, adj. 6 , 1398: “openne
and playne.” OD, 6 , 1729: “plain and easy to be
understood.” OD, 4, 1736: “plain and obvious.”
Jesp 11, 373, “A variant form of adjective-subjunct
is found in colloquial and dialectal speech when and
is added between the two adjectives.”
L 93 : WG 113 . I wish I was wonderful. LE
OD, s.v. be, vb. 7 fl, 1684-1787. Jesp IV,
129, cites Defoe, Swift, Fielding, Austen, Byron,
Marryat, Thackeray, Dickens, Hardy, Meredith,
Wilde, Norris, and others.
L 94 : WG no. I’ve no doubt but what he will
come. CE
OD, s.v. but, 30, 1662-1884, “dial, and
colloq.” The citations are colloquial rather than
dialect.
L 95 : WG 94 . What was the reason for Bennett
making that disturbance ? LE
Curme, Syntax, 488, 1338-1925. Hall
EU, 136-143, cites 217 instances in fifty-three au¬
thors.
L 96 : WG 150 . Can I be excused from this class? LE
OD Sup, vb . 1 B 6 b, 1894-1905, citing
Hardy. Curme, Syntax, 411.
38 FACTS ABOUT CURRENT ENGLISH USAGE
L 97 : WG 122 . Haven’t you got through yet? CE
OD, s.v. get, vb. 43, 1694-1895. Al¬
though no comment is made, the citations are chiefly
colloquial in nature.
L 98 : WG 78 . Everyone was here, but they all
went home early. LE
OD, s.v. every, 10 c, 1735-1877, “The
pronoun referring to everyone is often plural.”
L 100 : WG 90 . My folks sent me a check. CE
OD, 4, 1715-1833. The citations are
colloquial in nature. Webster, “colloquial.”
L 101 : WG 107 . He came around four o’clock. ALE
OD Sup, prep. 4 b, 1888-1920, “U.S.”
L 102 : WG 135 . If it had been us, we would admit
it. CE
OD, 5 d, 1883-1897, “Common in dia¬
lect and colloquial use, and occasionally employed
in writing.”
#
B. linguists, “disputable”; whole group, “established.”
L 108 : WG 60 . That clock must be £xed. ACE
OD, vb. 14 b, 1769-1891, “chiefly U.S.
colloq.” Jesp 111, 244, “in U.S. = repair.”
L 109 : WG 43 . My contention has been ^ proven
many times. LE
OD, s.v. prove, vb. A 2, 1536-1899,
“properly in passive.”
L hi: WG 16 . One rarely likes to do as he is told. LE
OD, pron. 21 , 1607-1652. Curme, Syn¬
tax, 531. “The older forms he, his, him, still linger
on.” Galsworthy cited.
L 112 : WG 55 . He never works evenings or Sun¬
days.
OD Sup, s.v. evening, d, 1862-1926,
“chiefly U.S. and dialect.”
ALE
THE "DISPUTABLE” USAGES
39
L 114 : WG 53 . The Rock Island depot burned
down last night.
OD, 5, 1830-1892, “U.S ”
L 118 : WG 56 . He went right home and told his
father.
OD, adv. 3 c, 1849-1901, “U.S ”
C. BOTH GROUPS, “DISPUTABLE.”
L no: WG 120 . Sam, who was then in town, was
with me the three or four £rst
days.
OD, s.v. first, adj. 2 c (a), 1340-1781,
“This still survives, though it is now rarely used
when numbers above 3 or 4 are concerned.” Webster,
“may otherwise follow numbers.”
L 113 : WG 98 . They have gotten a new car this
year.
* OD, 1868, “In U.S. literature gotten is
still very common.” Webster, “esp. in U.S.”
L 115 : WG 123 . Sitting in back of John, he said,
“Now guess what I have.”
OD does not record expression with in¬
troductory preposition in . Curme, Syntax, 564,
“popular American.”
L 116 : WG 125 . I took it to be they.
OD, 1 b, 1380-1890, “now only dial,
or illiterate.” This statement is made without refer¬
ence to the particular construction given here.
L 117 : WG 108 . I guess I’ll go to lunch.
OD, vb. 6 , 1692-1885, “colloquial in
Northern U.S.”
L 119 : WG 11 S. He could write as well or better
than I.
Not recorded.
ALE
ALE
LE
ALE
ACE
Dial
ACE
40 FACTS ABOUT CURRENT ENGLISH USAGE
L 120 : WG 144 . I expect he knows his subject. CE
OD, 6 , 1592-1877, “Now rare in liter¬
ary use.” Curiously enough five examples are cited
from the nineteenth century. The OD comments
further, “is often cited as an Americanism but is
very common in dialectal, vulgar, or carelessly
colloq. speech in England.” Webster “chiefly col¬
loquial.”
L 121 : WG 102 . I can't seem to get this problem
right. LE
OD does not record; Webster records
without comment.
L 122 : WG 126 . I was pretty mad about it. ACE
OD, 5, 1300-1867, “In many dialects
in Great Britain and the U.S. the ordinary word for
‘angry.’ ” OD Sup gives U.S. examples, colloquial
in character, 1887-1908.
L 123 : WG 157 . Either of these three roads is
good. * LE
OD, 4 c, 1616-1845.
L 124 : WG 159 . You are older than me. LE
OD, 6 b, 1606-1804. Shakespeare. Rich¬
ardson, Byron cited. Hall EU, 153, “It is found in
Shakespeare, Swift, Prior, Pope, Southey, and A. H.
Clough.”
L 125 : WG 178 . What are the chances of them be¬
ing found out? CE
Curme, Syntax, 489, “Often employed
in colloquial speech.” Cites Caxton and Latimer.
L 126 : WG 116 . There is a big woods behind the
house. CE
OD does not record. Webster, “More
often in pi. and often, chiefly in coll, use construed
as a singular.”
THE “DISPUTABLE” USAGES
41
L 127: WG 127. I know it to be he.
OD does not record he after the infini¬
tive of the verb to be.
L 128: WG 95. Do you v/ish for some ice cream? LE
OD, vb. 2 a, 1526-1867.
L 129: WG 171. Intoxication is when the brain is
affected by certain stimulants.
OD and Webster do not record.
L 130: WG 154. Neither of your reasons are really
valid. LE
OD, B 2 d, 1611-1875, citing Shake¬
speare, Dryden, Newman, and Ruskin.
L 131: WG 91. He dove off the pier. ALE
OD, 1882-1893, with comment, “U.S.
and Eng. dial.” Webster, “colloquial,” but Horwill,
Modern American Usage, 110, has twentieth-century
literary examples. The OD citations are literary in
character.
L 132: WG 142. Trollope’s novels have already be¬
gun to date. LE
OD Sup, vb. 2, 1895-1928.
L 133: WG 141. Will you go? Sure. Dial
OD, adv. B 3 c, 1813-1862, “Dial ”
L 134: WG 148. He is kind of silly, I think. CE
OD and Sup., s.v. kind, sb. 14 d, 1796—
1889, “colloq ”
L 135: WG 117. I will probably come a little late. LE
OD, s.v. will, vb. 1 16, 888-1923, with
comment, “since 17th c. almost exclusively in Scot¬
tish, Irish, provincial, or extra-British use.” While
some citations are dialect, many are literary. Jesp
IV, 256, “this usage is constantly gaining ground.”
Shelley and Shakespeare quoted. Curme, Syntax,
363, “In American colloquial speech will is now the
42
FACTS ABOUT CURRENT ENGLISH USAGE
more common form in the first person.” “Shall . . .
is still the preferred form in the higher grades of the
literary language in America, though not so uni¬
formly used as it once was.” Fries, “The Periphrastic
Future with Shall and Will” Publications of the
Modern Language Association, 40 (1925), 963 ff.,
“The first person with will has always predomi¬
nated.”
L 136 : WG 145 . That was the reason for me leav¬
ing school.
Curme, Syntax, 489, “often employed
in colloquial speech.” Latimer and Caxton cited.
L 138 : WG 155 . Ill swear that was him.
OD, 3, 1605-1840, “Common in collo¬
quial language from end of 16th century.” Shake¬
speare, Johnson, and Burke cited.
L 139 : WG 179 . Well, that’s going some.
OD, adv. C 2 c, 1866-1894, “U.S ” The
examples, however, are colloquial in character.
L 140 : WG 128 . Leave me alone or else get out.
OD, vb. 13, 1400-1885.
L 141 : WG 161 . Of two disputants, the warmest is
generally in the wrong.
Curme PSA, 188, “Sometimes in the
literary language,” cites Thoreau. Jesp II, 204,
“Found very frequently in good authors,” with
range of illustrations 1470-1896, including Shake¬
speare, Defoe, Stevenson. See also Hall EU, 280,
and an article by Russell Thomas, “The Use of the
Superlative Degree for the Comparative,” English
Journal, College Edition, 24 (1935), 821-829, where
over 75 examples of 23 adjectives from Literary
English, ranging in date from 950 to the present
time, are cited.
CE
CE
ACE
LE
LE
THE “DISPUTABLE” USAGES
43
L 142 : WG 182 . It was good and cold when I came
in.
OD Sup, s.v. good, adv. B d, 1834-
1926, “U.S. Colloq ”
L 143 : WG 156 . We haven't but a few left.
OD does not record.
L 144 : WG 119 . In the collision with a Packard,
our car naturally got the worse o£
it.
OD, sb. 4 b, 1205-1888; the citation
for 1860 is a phrase with of .
L 145 : WG 143 . I wouldn't have said that if I had
thought it would have shocked
her.
OD has no specific information on this
question of tense sequence.
L i 46 :*WG 164 . Yourself and your guests are in¬
vited.
OD (as simple subject) 4 a, 1400-
1799. Webster, “Archaic and dialect.” Hall EU, 175—
176 cites 37 authors from Malory to Stevenson as
using the self forms but does not differentiate as to
the various possible uses.
L 147 : WG 140 . The man was very amused.
OD, adv. B 2 c, 1641-1877.
L 148 : WG 86 . Such naif actions seem to me ab¬
surd.
This spelling recorded in OD 1598-
1885.
L 149 : WG 169 . It seems to be them.
K OD, B 3 b, 1654-1888, “Common col¬
loquially.”
ACE
LE
Dial
LE
LE
CE
44
FACTS ABOUT CURRENT ENGLISH USAGE
L 150 : WG 166 . Everybody bought their own
ticket. LE
OD, 1530-1871, Sidney and Ruskin
cited.
L 151 : WG 151 . Say, do you know who that is? ACE
OD, vb. 12 b, 1857-1888, “Colloq.
U.S”
L 152 : WG 162 . I suppose that’s Mm. CE
OD, 3, 1515-1840, Shakespeare and
Van Brugh cited. “Common in colloq. use from end
of 16th c”
L 153 : WG 133 . I can't help but eat it. LE
OD, s.v. help, vb. 11 b, 1894.
L 155 : WG 158 . There is a row of beds with a cur¬
tain between each bed. LE
Jesp II, 203, cites Shakespeare, Fielding,
and Dickens. Curme, although disapproving of the
construction, cites George Eliot.
L 156 : WG 136 . If I asked him he would likely re¬
fuse. ALE
OD, adv. 13 2, 1380-1895, with com¬
ment, “Rare exc. Scotch or dial.” Webster records
without comment. Horwill, Modern American Usage,
190, has several examples of this construction.
L 157 : WG 160 . John didn’t do so bad this time. ACE
OD Sup, 1816-1890, “U.S ” The ex¬
amples are chiefly colloquial in nature.
L 158 : WG 92 . Cities and villages are being
stripped of all they contain not
only, but often of their very in¬
habitants.
OD does not record.
L 159 : WG 80 . Everybody's else affairs are his
concern.
Curme PSA, 170, “older form.” Hall
LE
THE “DISPUTABLE” USAGES 45
EU, 87, cites instances of this construction in
Thackeray, Mark Twain, and Henry James.
L i6g: WG 165 . It don't make any difference what
you think. ACE
OD Sup, s.v. do, A 2 c y, 1670-1862,
labeled “American. 55 The examples are generally col¬
loquial in character.
L 161 : WG 182 . I read in the paper where a plane
was lost. CE
OD, 10 c, citations range from 1300-
1611 (two Shakespeare) and instances are given
from modern colloquial speech.
L 162 : WG 146 . That boy’s mischievous behavior
aggravates me. CE
OD, 7,1611-1858, “familiar . 55 Richard¬
son and Thackeray cited.
L 165 : WG 129 . Yes, our plan worked just £ne. Dial
OD, adv. C 1 b, 1385-1890, “Dial. 55
L 166 : WG 153 . The fire captain with his loyal
men were cheered. LE
Curme, Syntax, 49, “The plural is often
found here in older English and is sometimes still
used . 55 Cites Bible, I Tim. vi. 6 . Jesp II, 176, cites
Shakespeare, Bunyan, and Shaw.
L 168 : WG 137 . The British look at this differ¬
ently than we do. LE
OD, 1665-1844.
L 169 : WG 147 . Most anybody can do that. ACE
OD and Sup, adv. B 4 b, 1770-1904,
“Dial, and U.S. 55 Webster , “Dial, and colloq. 55
L 170 : WG 149 . It is liable to snow tonight. LE
OD, 3 b, 1682-1896. Horwill, Modern
American Usage, considers it an Americanism, but
the OD citations do not bear out his contention.
46 FACTS ABOUT CURRENT ENGLISH USAGE
L 171 : WG 168 . They went in search for the miss¬
ing child.
Not recorded.
L 173 : WG 138 . John was raised by his aunt. ALE
OD, vb . 1 10, 1744-1870, “Now chiefly
U.S.” Horwill, Modern American Usage cites ex¬
amples from contemporary American writing.
L 174 : WG 175 . Martha don't sew as well as she
used to, ACE
See evidence for L 160 above.
L 175 : WG 163 . He most always does what his
wife tells him. ACE
OD and Sup, adv. B 4 a, 1584-1901,
“Dial, and U.S.” Webster, “Dial, and colloq ”
L 177 : WG 180 . My experience on the farm helped
me some, of course. ACE
OD, adv. C 2 b, 1825-1889, “U.S.”
Horwill, Modern American Usage cites Lincoln Stef-*
fens, Webster, “colloq.”
L 178 : WG 152 , It’s real cold today. ALE
OD, adv. B, 1658-1887, “Chiefly Sc.
and U.S.”
L 179 : WG 172 . His presence was valueless not
only, but a hindrance as well.
Not recorded.
L 180 : WG 174 . We don’t often see sunsets like
they have in the tropics. LE
OD, adv. B 6 a, 1530-1886, “Gener¬
ally condemned as vulgar or slovenly though ex¬
amples may be found in many recent writers of
standing."
L 182 : WG 173 . She leaped off of the moving car. Dial
OD, adv. 7 b, 1593-1875, “Formerly
and still dialectically.” Shakespeare cited.
THE “DISPUTABLE” USAGES
47
L 184 : WG 170 . It is only a little ways farther. ALE
OD, sb . 1 IV 23 c, 1588-1907. Webster,
“Dial, and coll.” Horwill, Modern American Usage,
has examples from American writings.
L 185 : WG 177 . It looked like they meant business. ACE
OD and Sup, adv. B 6 e, 1493-1898,
“Dial, and U.S.” The U.S. examples are definitely
colloquial.
L 187 : WG 134 . The child was weak due to im¬
proper feeding. ALE
OD and Sup, a 9, 1661-1886, “Freq.
in U.S.”
D. linguists, “disputable”; whole group, “illiterate.”
L 137 : WG 184 . They eat [et] dinner at twelve
o’clock. CE
OD, 1835, is latest citation for this spell¬
ing, although H. C. Wyld in his Universal Diction¬
ary (London, 1931) considers eat the normal form
and ate as obsolescent. OD, H. W. Fowler, Modern
English Usage, and Daniel Jones, An English Pro¬
nouncing Dictionary, all of which record British
rather than American pronunciation, give [et] as
the more frequent form. Since this is a question of
pronunciation rather than spelling, it is classed as
a colloquial usage.
L 154 : WG 185 . Aren't I right? CE
Curme PSA, 248, gives twentieth-cen¬
tury examples and classes as colloquial.
L 163 : WG 192 . The stock market collapse left me
busted. ACE
OD Sup, 1881-1920, “U.S. colloq ”
L 164 : WG 195 . Neither author nor publisher are
subject to censorship. LE
OD, Aid, 1759-1874. Johnson, Cow-
per, Southey, and Ruskin cited.
48
FACTS ABOUT CURRENT ENGLISH USAGE
L 167 : WG 198 . Don’t get these kind of gloves. CE
OD, sb. 14 by 1564-1797, “common
colloquially.” Also sort with these or those, 1551-
1872.
L 172 : WG xgi. I suppose I’m wrong, ain't I? CE
OD, latest citation, 1799, with com¬
ment: “Popular dialect of London and elsewhere.”
Webster, “dialect.” Curme PSA, 248, classifies it as
colloquial, and the Illustrative citations are certainly
not dialect.
L 176 : WG 187 . It sure was good to see Uncle
Charles. ACE
OD and Sup, adv. B 3, 1425-1913,
“Dialect, Irish, and U.S.” Citations include Sidney,
Milton, and Dryden.
L 181 : WG 203 . I am older than Mm. CE
OD, 3, 1759-1764, but the predicative
him citations range from 1381-1840. “Common in
colloquial language from the end of the sixteenth
century.”
L 183 : WG 186 . She sung very well. LE
OD citations up to 1877. Webster
records without comment.
L 186 : WG 190 . Do it like he tells you. LE
OD, 1530-1886, “Now generally con¬
demned as vulgar or slovenly, though examples may
be found in many recent writers of standing,” Cita¬
tions include Shakespeare, Southey, and William
Morris.
E. linguists, “illiterate”; whole group, “disputable.”
L 191 : WG 181 . The dessert was made with whip
cream.
Not recorded.
THE “DISPUTABLE” USAGES
49
L 192 : WG 180 . Now just where are we at? ALE
OD Sup, s.v. at, prep. 1 d, 1859-1914,
“U.S.”
L 193 : WG 167 . The kitten mews whenever it
wants in. ACE
OD, s.v. want, vb. 4 f, 1844-1897,
“Scotch, North Irish, U.S. colloquial.”
L 196 : WG 176 . Reverend Jones will preach. LE
OD, 2 c, last citation 1657, “In early
use without the? 3 E. C. Ehrensperger in American
Speech, October, 1931, 41, “In this country the
practice of omitting the is widespread . . . the
Boston Transcript, the Harvard Alumni Bulletin,
many church periodicals, etc., omit it.”
L 204 : WG 139 . The data is often inaccurate. ALE
OD, no information. Webster, “Not in¬
frequently used as singular.” Horwill, Modern Amer¬
ican Usage, 96, cites Professors E. J. Haskin and
C. McCarthy, and Col. House,
The results of this survey of the “disputable” expressions may
be most conveniently presented in tabular form. The 121 items
belonging to this group were found to be distributed among the
various categories as shown on page 50.
Undoubtedly, the most striking feature of this tabulation is
the high proportion of these expressions which are recorded as
occurring in Standard English, written and spoken. Adding the
fifty items which are recorded as belonging to Literary English,
the thirteen to be found in American Literary usage, the twenty-
five colloquial expressions and the eighteen American Colloquial,
we find that 106 of the 121 items, which according to a survey
of opinion seemed to be disputable, are, on the basis of recorded
fact, actually in cultivated use today. This is a proportion of
87 per cent. Of the remaining fifteen items, six were in standard
use at some previous time.
50
FACTS ABOUT CURRENT ENGLISH USAGE
TABLE II
Status op Disputable Expressions in Recorded Usage
C. E. U. Ranking
LE |
ALE
CE
ACE
Dial
Arch
NR
Total
A. Linguists ‘‘estab¬
lished” Whole Gr.
“disputable” .
23
1
10
1
1
36
B. Linguists “disputa¬
ble” Whole Gr.
“established”.
2
3
1
6
C. Both Gr. “disputa¬
ble” .
21
7
10
13
5
8
64
D. Linguists “disputa¬
ble” Whole Gr. “il¬
literate” .
3
5 !
2
10
E. Linguists “illiterate”
Whole Gr. “disputa¬
ble” .
1
2
i
1
_
1
5
Total.
50 ;
13
25 :
18 '
6
0
9
121
Legend: LE —Literary English Dial — Dialect .
ALE —American Literary English Arch —Archaic
CE —Colloquial English NR —Not Recorded
ACE —American Colloquial English
We may conclude, then, first of all, that the teacher is not
only safe in accepting the so-called “established” usages of the
Leonard report, but there are seven chances out of eight that a
“disputable” item is wholly current in standard English as well.
Nor is there, from the evidence, any reason to suspect that these
106 items are to be considered particularly inelegant. In other
words, the teacher may advise his pupils to avoid the “disputa¬
ble” usages if he wishes to. That is his privilege. But his censure
of these expressions cannot be on the basis that they are not to
be found at present on the pages of reputable writers or in the
mouths of cultivated speakers.
To the student of language this tabulation will demonstrate
also how much more conservative a survey of opinion about
language is apt to be than the facts of the language actually
THE “DISPUTABLE” USAGES SI
warrant. The Introduction to the grammar section of the Leon¬
ard report, maintained that dictionaries, because their citations
were drawn from literary examples of acknowledged value, were
necessarily slower and more conservative than usage Itself, that
the dictionary method, valid though it might be, must of neces¬
sity result in a lag of several years between the adoption of a
given usage and its appearance in a dictionary. 17
There are here two mistaken assumptions, one of which is
expressly stated and the other implied. That the dictionary
record of fact does not lag behind opinion, but on the contrary
is well in advance of it, has been clearly demonstrated and needs
no amplification. The implication that these disputable expres¬
sions are so new that they have not yet had time to be recorded
in the dictionaries is likewise not borne out by the facts. Only
nine of the 121 disputable expressions were not recorded in any
of the sources used, and two of these nine (Nos. 158 and 179)
are infelicities in style rather than matters of grammar. The
Oxford Dictionary alone recorded all but twenty-three of the
121 . Of the whole group of 121 disputable expressions, twenty-
seven are recorded as arising in the nineteenth century, ten in
the eighteenth, twenty In the seventeenth, twenty-two In the
sixteenth, and twenty-four sometime before 1500; that is, either
in the Middle or Old English periods. In other words, the ex¬
pressions about which puristic objections center are not so much
neologisms as they are old forms and usages of the language
which are struggling to survive.
Finally, It is evident that this analysis should dispose once
and for all of the journalistic cry of heresy and radicalism so
frequently raised against the Leonard report. A survey of fact
rather than of opinion would, in all probability, have increased
the number of established usages from a meager seventy-one to
177.
17 Current English Usage, p. 95. Reprinted in this monograph as page 65.
V <?£L
THE “ILLITERATE 55 USAGES
Before considering in detail the thirty-eight items which were
voted “illiterate” by both groups of judges, it will be helpful
to refresh our memory as to the implications of this term as it
was used in the instructions to the judges. In the ballot, the
fourth category was defined in the following terms:
4. Popular or illiterate speech, not used by persons who wish to
pass as cultivated, save to represent uneducated speech, or to be
jocose; here taken to include slang or argot, and dialect ^forms not
admissible to the standard or cultivated area; usually called “vulgar
English, 55 but with no implication necessarily of the current meaning
of vulgar: “naif, popular or uncultivated English. 55
It is unfortunate that when the results of the survey were put
together the single word illiterate was chosen as a defining label
to represent this whole group. In a country such as ours, where
literacy is a virtual pre-requisite for respectability, the term
illiterate serves to cast a gratuitous opprobrium upon any ex¬
pression or group of expressions to which it is applied. Nor does
it adequately sum up what is indicated by the detailed explana¬
tion quoted above. Certainly slang is not an “illiterate 55 form of
expression; its use implies nothing about literacy, one way or
another. Regional dialect forms may be used by speakers about
whose cultivation there is no question. In terms of the definition
given to the judges, “non-standard 55 would have been a much
more accurate label. This term includes within its scope slang
m
Summary Sheet of Ballots—Grammatical Usage Study
s ^^^^^^ 1 ^rtrHrHC'10*3C<lC?C<IC^C , s|CaC
•^H\D(MOHt?CM>in£^VOeCC
iWrt^'OHN^co^coo;
I
< gq frj Oft 00
I00\00
THE “ILLITERATE” USAGES
S3
as well as regional and social dialects. None of these are Standard
English. However the term “non-standard” says nothing more
about these usages than just that, nor does it suggest anything
about the social status, cultivation, taste, or literacy of those who
use them.
Let us turn now to an examination of the factual sources for
the evidence concerning those expressions which the judges con¬
sidered beyond the pale of Standard English:
L 188 : WG 199 . John had awoken much earlier
than usual. LE
OD, s.v. awake and wake, 1633-1924,
but comments that woken seems obsolescent. Cur me
PSA, 306, cites Walpole and London Times, 1932,
and comments, “In British English the strong forms
awoke, awoken, woke, woken still occur in the past
participle. In American English . . . confined to
colloquial and popular speech.” Webster, s.v. awake
labels awoken “obs.” but s.v. wake comments “some¬
times woken.”
L 189 : WG 209 . I haven't hardly any money. ^ Arch
OD, 7, “Formerly sometimes (as still
in vulgar use) with superfluous negative.” One cita¬
tion, 1674.
L 190 : WG 196 . The engine was hitting good this
morning. Arch
OD, adv. B a, 13. .-1887, “Obs., fare
except in vulgar or slang phrases.”
L 194 : WG 193 . A woman whom I know was my
friend spoke next. LE
OD, 11 , 1467-1906, Dickens cited, but
comments, “Used ungrammatically.” Jesp III, 198,
“The idiom is found in many authors of repute.”
Chaucer, Caxton, Shakespeare, Boswell, Shelley,
Keats, Kingsley, Kipling, Galsworthy, Wells, Drei¬
ser, and numerous others cited.
Arch
54 FACTS ABOUT CURRENT ENGLISH USAGE
L 195 : WG 206 . He drunk too much ice water.
OD, 1300-1648, “drunk began to re¬
appear for sing, as well as pi. at end of 16th c. and
is occasional to 19th.” Webster, “formerly also
drunk”
L 197 : WG 207 . All came except she.
OD, 4, 1530-1881, “Now rare.” Shake¬
speare cited three times.
L 198 : WG 189 . The party who wrote that was a
scholar.
OD, 14, 1541-1888, “Formerly com¬
mon and in serious use; now shoppy, vulgar, or
jocular.” Jonson cited. Webster, “slang.”
L 199 : WG 197 . My Uncle John, he told me a
story.
OD, 3, 1000-1839, all the citations ex¬
cept one are poetic. “Common in ballad style and
now in illiterate speech.”
L 200: WG 216. He begun to make excuses.
OD, 1563-1793, Pope cited. “An alter¬
native from the old plural begun has also come down
to the present day.” Curme PSA, 307, marks it as
an older literary form and comments that it survives
in popular speech.
L 201: WG 200. I calculate to go soon.
OD and Sup, 7, 1822-1859, “U.S. col-
loq ” Webster, “Colloq. U.S.”
L 202: WG 188. This is all the further I can read.
Although Current English Usage says
that this expression is not recorded in any dictiona¬
ries, OD, s.v. all, II 3 c, does give a group of con¬
structions from which this undoubtedly arose. The
citations range from 1250-1633.
L 203 : WG 217 . That ain't so.
OD, 1778-1865, “Popular dialect of
London and elsewhere.” Webster, “Dial, or illit.”
Arch
Arch
Arch
CE
ACE
Arch
Dial
THE “ILLITERATE” USAGES
L 205 : WG 214 . He looked at me and says , . .
OD, vb . 1 3 b 1682-1887, “In this use,
the 3d sing, pres, is often substituted colloq. for
the pa. t. said ” Jesp IV , 19, “Very often this present
tense (historic) alternates with the preterit.” Gam¬
mer Gurton’s Needle, Shakespeare (5 times), Bible,
Defoe, Shaw cited. 1550-1903.
L 206 : WG 213. I must go and lay down.
OD, vb . 1 43, 1300-1900, “Now (except
in nautical language) it is only dialectical or an
illiterate substitute for lie . . . . In the 17th and 18th
centuries, it was not app. regarded as a solecism.”
Fielding and Byron cited.
L 207 : WG 210 . Ain't that just like a man.
See evidence under L 203.
L 208 : WG 194 . Both leaves o£ the drawbridge
raise at once.
OD, vb . 1 35, 1470-1761. OD Sup has
later U.S. examples from reputable writers, 1770-
1911.
L 209: WG 202. The people which were here have
all gone.
OD, 9, 1300-1703, “Now only dial, ex¬
cept in speaking of people in a body.” Webster,
“Arch, and dial.”
L 210: WG 204. I have drank all my milk.
OD, 1704-1819, “From 17th to 19th
c. drank was intruded into the pa. pple., prob.
to avoid the inebriate associations of drunk/ 3
L 211: WG 221. That there rooster is a fighter.
OD, 2 c, 1742-1863, “Dial, and vulgar.”
Webster, “Dial, and illiterate.”
L 212: WG 183. The old poodle was to no sense
agreeable.
This item was generally misinterpreted
by both compilers and judges. The vote was con-
55
CE
Dial
Dial
ALE
Dial
Arch
Dial
56
FACTS ABOUT CURRENT ENGLISH USAGE
sidered valueless at the time the Leonard report was
written. See Current English Usage , 141, note 3 (p.
108 of the monograph). For this reason, it has been
excluded from consideration here.
L 213 : WG 205 . One of my brothers were helping
me.
OD and Webster do not record.
L 214 : WG 201 . I enjoy wandering among a li¬
brary. LE
OD, prep. A 2 , 1175-1810, with collec¬
tives and mass words. The test sentence is from De
Quincey. Webster, “Regularly followed by a plural
or collective noun.”
L 215 : WG 211 . A light complected girl passed. ACE
OD, 1860-1873, “U.S. dial, or colloq.”
OD Sup, has further examples, 1822-1906. Webster,
“Dial., U.S” Horwill, Modern American Usage,
78, “American colloquial.”
L 216: WG 208. I want for you to come at once. *
Not recorded in any of the sources.
L 217: WG 226. He won’t leave me come in. ACE
OD Sup, vb. 3 e, 1840-1910, “U.S.
Colloq.”
L 218: WG 228. There was a orange in the dish.
Not recorded.
L 219: WG 218. It was dark when he come in.
OD, 855-1888, “Hardly appears after
1500 in the literary language though still widely
prevalent in Midland and Southern dialects.” The
OD treatment makes it difficult to separate the come
spellings with phonetic values of [kom] and [Lim].
L, 220: WG 222. You was mistaken about that,
John.
OD, s.v. be, vb. Ill 6.2 f, 1340-1837,
“Still dial, in all persons.” Webster, “Widely used
Dial
Dial
THE “ILLITERATE” USAGES
57
in the 18th century, often by standard authors,
now regarded as grammatically incorrect or illiter¬
ate.”
L 221 : WG 229 . I wish he hadn't of come.
This involves two factors, the redundant
use of have, had, in compound tenses and the un¬
stressed of [ov] variant for have. Concerning the
first, OD comments, s.v. have, vb. 26, “In 15th and
16 c. occur many instances of redundant have, had,
in the compound tenses.” Citations with have after
had, 1470-1677, while OD Sup cites U.S. examples
1816-1911. Concerning of for have, OD Sup, s.v. of,
comments, “U.S. dial, or coll, variant,” and has
citations 1847-1916.
L 222: WG 212. Hadn’t you ought to ask your
mother?
OD, s.v. ought, vb. IV 7 c, 1836-1895,
“Mod. Dial.” Jesp IV, 128, cites Bennett, Winston
•Churchill, Ade, and several examples from Sinclair
Lewis. The examples in Jespersen appear to be col¬
loquial rather than dialect.
L 223: WG 227. My cold wa’nt any better next day.
OD, s.v. he, III 6 ft, and s.v. wa’n’t,
1702-1865, “Dialectically were, war (sg.) occur,
hence the negative warn’t, wa’n’t in 18th c. drama¬
tists.” Webster, “Dial.”
L 224 : WG 223 . If John had of come, I needn’t
have.
See evidence for L 221.
L 225: WG 219. I had hardly laid down when the
phone rang.
See evidence for L 206.
L 226: WG 230. He did noble .
Not recorded.
ACE
CE
Dial
ACE
Dial
58 FACTS ABOUT CURRENT ENGLISH USAGE
L 227 : WG 224 . Somebody run past just as I
opened the door. Dial
OD, 1382-1869, “Dial.” Webster,
"Dial.”
L 228 : WG 225 . Just set down and rest awhile. Dial
OD, 5, 1205-1897, set down, s.v. set,
vb. 143 h ( b ), 1400-1809, “Dial. or vulgar . 77 Web¬
ster, “Now illit. and dial . 77
L 229 : WG 220 . The neighbors took turns setting
up with him. Dial
See evidence for L 228.
L 230 : WG 215 . They swang their partners in the
reel. LE
OD, 1000-1912, “rarely swang ” Curme
PSA, 318, marks it as older literary form. Webster,
“Archaic past tense . 77
The results of this survey of the “illiterate” or non-standard
expressions may be conveniently summarized in the following
table:
TABLE III
Status or Illiterate Expressions in Recorded Usage
Literary English. 4 Dialect . 12
American Literary English ...... 1 Archaic. 8
Colloquial English. 3 Not recorded. 5
American Colloquial English .... 5
In interpreting the factual record of the “illiterate” expres¬
sions, we are faced with a situation somewhat different from that
which prevailed with the disputable group. The “disputable 77
expressions, it will be recalled, were those concerning which a
survey of opinion showed such divergence that a supplementary
survey of fact was imperative if we were to come to any opinion
about them. As to the thirty-eight items just examined, however,
there was little or no lack of unanimity on the part of the judges.
THE “ILLITERATE” USAGES 59
All agreed fairly well in consigning them to the limbo of the
illiterate. Because of this, the system of classification employed
previously was left unchanged. That is to say, when an expres¬
sion was labeled in the dictionaries as “dialect and illiterate,”
it was classified here as dialect; when it was labeled “illiterate”
in present use but had obviously been in current use at an earlier
period, it was classified here as “archaic.” This procedure was
followed in the hope that it might lead us to some useful con¬
clusions about the nature of these non-standard expressions.
First of all, it should be noticed that thirteen of the thirty-
eight items, roughly one-third, are recorded in reputable literary
or colloquial use, either in England or America. Again the ex¬
treme conservatism of opinion about usage, as compared with
the factual record of usage itself, is strikingly demonstrated.
It is also noteworthy that all but five of the thirty-eight items
were recorded in the sources employed, and furthermore, that
all thirty-three of those for which any record was found ap¬
peared In the Oxford Dictionary or its supplement. Thus we
find that not only the disputable but even most of these con¬
demned expressions were in accepted usage at some former
period, but that over half of them (the twelve dialectal and the
eight archaic) are now confined to particular regional or social
dialects, that is to say, to limited, non-standard spheres of usage.
We are reminded again how much of non-standard, “incorrect,”
or questionable language has a continuous history and tradition
behind it; it is not created on the spur of the moment but, to
indulge in a simile, is like an underground stream which pops
up into the light of day where it is least expected and frequently
not welcome.
There is one final observation to be made in connection with
this group of expressions. This is in connection with the type of
error or supposed error condemned as illiterate. According to the
classification employed in the body of Current English Usage,
the thirty-eight illiterate items represented sixteen different
60
FACTS ABOUT CURRENT ENGLISH USAGE
types of errors. Yet no less than twenty of the thirty-eight ex¬
pressions were concentrated in those categories which dealt with
the forms and uses of verbs. There were eight items which were
concerned with past tense forms alone. Of the twenty expres¬
sions involving questions of verb form and use, seven were
found to be Standard English, two archaic, and the remaining
eleven were dialectal forms.
All of this would seem to suggest that in respect to matters
of verb formation and use, our prejudices are heightened. This
is particularly true of the past tenses of strong verbs. As a
matter of fact, even the past tense form sung, given equal rank
with sang by all of the factual sources, was rated as Literary
English by two English teachers, as Colloquial English by four¬
teen, and as Illiterate by sixteen. This particular verb eventually
ended up in the “disputable 77 category, but there is no over¬
looking the fact that SO per cent of the judges considered a fully
accepted verb form to be illiterate. The past participle awoken,
still in literary use in England, was voted “illiterate , 77 and it is
particularly ironic that the very last item on the list, the most
discredited, was the form swang, not at all uncommon in British
speech.
■QrS YI t^SL
CONCLUSION
It is the earnest hope of both of the authors of this monograph
that their attitude toward the Leonard study will not be mis¬
understood or misconstrued. There is no question in our minds
as to the worth or significance of this pioneer work. We believe
that it has accomplished much in the cause of linguistic liberal¬
ism. It has set forth an enlightened attitude toward language
problems in a fashion that can be grasped by teacher and layman
alike. It has undoubtedly influenced to a considerable extent the
language textbooks which have been written and revised within
the past few years. It is not, however, a final or definitive piece
of work, and to consider it so would be a betrayal of the spirit
in which it was conceived.
We trust as well that our attitude and procedure in the present
study will not be misunderstood. We have considered it our
function to supplement Leonard’s work by displaying side by
side with the opinions he collected, the record of usage of each
item included in the original study. It has been our aim to be as
objective as possible. The classifications of the various items are
not to be taken as representing the opinions or the recommenda¬
tions of the authors. We have endeavored to let the facts of the
language speak for themselves, or at least those facts which ap¬
pear in the most reliable and scholarly treatments of Modern
English.
61
CURRENT USAGE IN GRAMMAR
REPRINTED FROM
CURRENT ENGLISH USAGE
By STERLING A. LEONARD
X-fe I <?£L
INTRODUCTION
The following brief statement of aims and methods is in¬
tended as an aid in the use and interpretation of the discussions
and tables that follow.
Up to the present, almost the only authoritative statements
of acceptable practice in English usage have had to be sought
in dictionaries and handbooks. Dictionaries have as their prime
function the recording of usage, but by their very nature most
of their citations have to be drawn from literary examples of
acknowledged value; this method, valid though it may be, must
of necessity result in a lag of several years between the adoption
of a given usage and its appearance in the dictionary. It has been
shown that most handbooks are based on traditional pronounce¬
ments of dubious value , 1 but even where this is not the case, the
value of the handbook is limited by the same consideration of
time that handicaps the dictionary. Since—as the following
study should make evident—allowable usage is based on the
actual practice of cultivated people rather than on rules of
syntax or logic, it seems desirable that some method be found
whereby this practice can be ascertained and made available for
reference. This study, it is hoped, constitutes at least a beginning
of a research that, to be useful, should be constantly pursued in
order that current usage may be placed on record.
The conclusions arrived at in the following pages were de-
1 S. A. Leonard: The Doctrine of Correctness in English Usage, 1700-1800.
University of Wisconsin Studies in Language and Literature, No. 25. Madison,
1929.
65
66
CURRENT ENGLISH USAGE
rived from a study of the results of two ballots. The first con¬
tained 102 expressions (indicated on Summary Sheet) of whose
standing there might be some question. This ballot was sub¬
mitted to a number of groups of judges whose standing quali¬
fied them to indicate what seemed to them to be the norm of
usage among educated people generally. The first group of
judges comprised a number of the foremost linguistic experts
in the world—lexicographers, philologists, and grammarians. As
trained observers of language ways, they were naturally quali¬
fied above all others to estimate the standing in actual cultivated
use of the various items on the ballot. Therefore, in the follow¬
ing discussion of the separate items, their comments are given
special prominence. Each item is followed by a number which
indicates the rank assigned it by the votes of this group, though
where the other groups show any significant divergence from the
judgment of the linguists, the fact is noted. The second group
consisted of active members of the National Council of Teachers
of English. A third group was composed of well known authors;
a fourth, of the editors of influential publications; a fifth, of
leading business men; a sixth, of members of the Modern Lan¬
guage Association; and a seventh, of teachers of speech. Returns
were received from 229 judges altogether. They should con¬
stitute a significant sampling of cultivated usage.
Following are the instructions to the judges as they appeared
on the ballot:
The following list of expressions represents an attempt to present
one or more examples from each of the levels or regions of usage sug¬
gested by Dr. Murray in the preface to the New English Dictionary.
We hope by getting a consensus of expert opinion on the classification
of these expressions to clarify and define more precisely the categories
themselves. We shall be grateful if you will cooperate by placing in
the blank to the left of each expression a number to correspond with
one of the tentative definitions following. The word or phrase about
which there is question of placement is underlined; no other part of
INTRODUCTION
67
the sentence which may perhaps belong to a different level should
influence a judgment as to the crucial expression. The problem of
pronunciation does not enter.
Score, please, according to your observation of what is actual usage
rather than your opinion of what usage should be. For example, if
you detest like as a conjunction, but observe it as a standard literary
use, you should mark it 1. Comments on any or all the expressions
or on reasons for your placements will of course be most welcome.
Finally, please do not mark according to your own definitions of
the categories or terms below—though we should be greatly helped
if you cared to send in such definitions also; but for the purposes of
this study use the definitions offered here, since all findings will have
to be understood in the light of these:
Key Number Definitions of Terms
1. Formally correct English, appropriate chiefly for serious and
important occasions, whether in speech or writing; usually
called “Literary English.”
2. Fully acceptable English for informal conversation, correspond¬
ence,* and all other writing of well-bred ease; not wholly ap¬
propriate for occasions of literary dignity: “standard, cultivated,
colloquial English.”
3. Commercial, foreign, scientific, or other technical uses, limited
in area of comprehensibility; not used outside their particular
area by cultivated speakers: “trade or technical English.”
4. Popular or illiterate speech, not used by persons who wish to
pass as cultivated, save to represent uneducated speech, or to
be jocose; here taken to include slang or argot, and dialect
forms not admissible to the standard or cultivated area; usually
called “vulgar English,” but with no implication necessarily of
the current meaning of vulgar: “naif, popular, or uncultivated
English.”
In tabulating and evaluating the ratings on the ballots, marks
of either 1 or 2 were taken as an indication that the expression
in question is considered allowable for use by educated people,
whether formally or colloquially. In certain instances, marks
of 5 or higher expressed special reprobation. These were not
68 CURRENT ENGLISH USAGE
numerous enough to make It worth while to give them special
weights In the tabulations.
Ballot II consisted of 130 additional expressions of the same
nature as those in Ballot I. This ballot, with the same instruc¬
tions, was submitted to substantially the same list of judges in
the groups of linguists and members of the National Council as
received the first ballot. The other groups of judges were not
asked to mark this ballot, and fewer judges in the groups named
made returns—to be specific, returns were received from 17
linguists and 32 members of the Council. In view of the fact
that these judges represented a highly selected group in so far
as their qualifications were concerned, while they were quite
unselected in regard to the shades of opinion they might have on
matters of usage, their votes are probably significant in indicat¬
ing the levels of usage to which the items on Ballot II at present
belong. Because the judges marking Ballot II were fewer, per¬
haps the conclusions drawn from this ballot are slightly less
reliable than those from Ballot I.
■Q^) ii gQ.
JUDGES 5 DISCUSSION OF SPECIFIC
ITEMS OF USAGE
The specific grammatical usages contained in the question-
aries are grouped here for ready reference according to their
grammatical classification, each item being accompanied by a
general summary of the judges’ estimate of its acceptability,
and by any especially illuminating comment thereon.
Each item Is followed by one of the following words as an
indication of its probable usage status: established, disputable,
illiterate; and by a number indicating its rank among the 230
usages, according to the linguists. The rank indicating most
complete approval is one; that indicating most complete dis¬
approval is 230. The ranking by the linguists is here given, as
the expertness of this group of judges makes their opinion most
significant. (For a list of the items arranged in the order of their
acceptability, see pages 4-11.) Items marked established have
been approved for literary or good colloquial use by at least
75 per cent of the judges and disapproved by not more than
25 per cent. Disputable items have been approved by fewer
than 75 per cent and disapproved by more than 25 per cent.
Illiterate items have been disapproved by more than 75 per cent
of the judges and approved by fewer than 25 per cent.
The numbering of the headings in the following presentation
has been so arranged that the arabic numerals always designate
the specific examples of usage considered by the judges. The
slight resulting inconsistency in the numbering scheme for cer-
69
70 CURRENT ENGLISH USAGE
tain sections Is more than balanced by the greater ease in locat¬
ing usage Items.
Nouns
I. Number
1. There is a large works near the bridge. ( 29 : established)
The English teachers ranked this much lower than did the linguists,
placing it in the class of disputable usages. However, only 2 of 17
linguists disapproved it.
2 . There Is a big woods behind the house. ( 126 : disputable)
British linguists remark that this expression is strange to them,
though, as one says, u a big works Is familiar.”
In Informal speech, this Is probably acceptable in the United States.
3 . The data is often inaccurate. ( 204 : illiterate)
There was more disagreement among the various groups of judges
on this item than on any other on either ballot. Possibly an additional
jury of scientists would have added further to the confusion. Speech
teachers (with only one judge among them disapproving), ranked
this as 7 out of 100; business men ranked it 25, authors 38, and so
on down to linguists, who ranked it 92. Uniformly with the other
items, this expression is placed among illiterate usages on the basis
of this last ranking; a composite judgment would have placed it
high among disputable usages.
All that can be said definitely is that there is no dictionary justifi¬
cation for the singular use of this plural form, and that such use,
from the point of view of cultivated usage, is still dubious, to say the
least.
II. Case
1 . Pikes Peak is in Colorado . 2 ( 105 : established as technical)
Many of the judges were puzzled by this expression. One says: “I
2 See also the conclusions as to this usage in the punctuation study, Current
English Usage, p. 54.
DISCUSSION OF SPECIFIC ITEMS 71
don’t understand what question arises here.” Another asks: “Is Pikes
here a misprint?”
This confusion possibly accounts for the relatively low rating of
this expression by the judges, and especially by the linguists, who
placed it lower than did any other group. Every other group yielded
a majority for approval. The Standard Dictionary spells the name
with the apostrophe; the New International spells it without.
Perhaps the two remarks (by linguists) which follow get at the root
of the question:
“What was the trouble with this sentence? If it is the apostrophe,
I should like to see that go. Has it not gone in many of the English
names—Kings Warwick, and so on? That use of the possessive which
ceases to be a possessive and sinks gracefully into the thing possessed
—apparently a spiritual symbol of the devouring nature of all pos¬
sessions ! ”
“A proper name of a place is determined by local usage.”
There are, of course, many other instances of the suppression of an
apostrophe in place names; for example, Teachers College, Citizens
National Bank, etc. Both the spelling and pronunciation of place
names properly follow local usage.
2. It is only a little ways farther. (184: disputable)
Judges and dictionaries agree, for the most part, that in spite of
the historical justification for this form, it is now dialectal. The vote
was about two to one for its inclusion among uncultivated usages.
Pronouns
I. Classes
a. INTENSIVE
i. They invited my friends and myself . (91: established)
One linguist says: “It occurs to me that I am willing to make an
exception of Omar’s ‘Myself when young,’ because of its sheer charm.
But I would shut it out everywhere else save for emphasis.”
The editors rated this highest, linguists second, business men last.
62 per cent of all the judges approved it, thus placing it low among
established usages. This would suggest that, while perhaps people
72
CURRENT ENGLISH USAGE
who are especially careful of their speech would avoid this expression,
nevertheless it would hardly be safe to condemn it as incorrect.
2 . Yourself and your guests are invited. ( 146 : disputable)
This expression is approved by a majority of the linguists, disap¬
proved by a large majority of the English teachers. It does not rank
with the similar use of myself.
b. IMPERSONAL PERSONALS
x. It says in the book that . . . ( 62 : established)
Although the English teachers rated this expression much lower
than did the linguists, its right to be considered established in good
colloquial usage is made quite clear by the linguists’ large majority
vote.
2. You had to have property to vote, in the eighteenth cen¬
tury. (76: established)
75 per cent of the judges approved this as good colloquial English;
12 per cent in addition aproved it as literary. The case Jor its in¬
clusion among established usages is perfectly clear.
3. They had numerous strikes in England in i860. (85: es¬
tablished)
This indefinite use of the pronoun they is approved as colloquial
by five out of six of the judges.
C. RELATIVES
This is a man ... I used to know. (Omitted relative) (9:
established)
Only eleven of more than 200 judges condemned the omission of
the relative by placing this expression in class 4. Nearly half of the
judges approved this by placing it in class 1 (formal literary English);
many of those approving it as colloquial English indicated that they
believed it verged on literary English. It may be of interest to note
that the editors placed this lower (rank 32) than did any other of
the groups of judges, although even they accepted it as established . 3
8 See Jespersen’s Grammar, Article on Hypotaxis.
DISCUSSION OF SPECIFIC ITEMS
73
Certainly no possible justification can be found for the practice
some teachers still pursue, of requiring children always to insert the
relative in such sentences.
II. Gender
1. Each person should of course bear his or her share of the
expense, (n: established)
One linguist says: “I prefer simply his. This seems to be a matter
of pleasing the women.”
Another says: “Correct but not commendable.”
This expression is correct but stilted. See note on item 1 under
Articles in this chapter.
2. This is the chapter whose contents cause most discussion.
(19: established)
None of the linguists, and only six (of 32) English teachers, con¬
demned this as illiterate. The rest of the judges approved it. To in¬
sist that students use a phrase such as of which when speaking of
inanimate objects is pedantic.
3. The people which were here have all gone. (209: illiterate)
“Which” is no longer in good standing when used to refer to people.
III. Number
1. None of them are here. (37: established)
Of the seven groups of judges, four are composed of teachers—
linguists, English teachers, members of the M. L. A., and speech
teachers; the other three are non-academic—authors, business men,
and editors. It is worthy of note that the four groups of teachers
placed this expression in the class of established usages, while the
others considered it disputable. That is, 80 per cent of the former
approved the expression, as against only 50 per cent of the latter.
Among the authors the rank was 40; among the linguists, 12.
An editor, defending his estimate of the sentence as vulgar, says:
“ ‘None of them are here’ is not correct. It is perfectly correct to say
‘None are here. 7 But ‘None of them 7 must mean [sic] ‘no one of
74 CURRENT ENGLISH USAGE
them.’ Hence it must be followed by the singular form of the verb.”
The groups of teachers find justification for their estimate in those
dictionaries which record usage. The N. E. D. says: “In later use
none commonly with plural verb.” (The earliest plural example
recorded in this dictionary is of 1580.) The Standard says: “When
the singular or plural equally well expresses the sense, the plural is
commonly used.” The New International testifies that “as subject,
none with a plural verb is the common construction.”
One linguist states: “The objection to none are never had any basis
in good usage.” An author says: “It is pure priggishness to pretend
that none is always singular.”
No authority can be found for condemning the use of “none” with
plural verb. See also kind ... are and one . . . were, in this same
chapter under Verbs IV, items 1 and 3 .
2. Everyone was here, but they all went home early. (98: es¬
tablished)
A business man remarks: “Everyone is synonymous with all, hav¬
ing a collective sense, and it seems to me quite permissible that the
word they may follow.”
The N. E. D. says: “The pronoun referring to everyone is often
pi.; the absence of a sing, pronoun of common gender rendering this
violation of grammatical concord sometimes necessary.”
138 out of 200 judges approve this as colloquial usage; each group
except that of the editors (where the votes are evenly divided) gives
a majority for 1 or 2 (acceptable) as against 4 (uncultivated). Cf.
neither, item 4 below.
3 . Everybody bought their own ticket. ( 150 : disputable)
Citations: Sam. Johnson: “'Everyone sacrifices a Cow or more,
according to their different degrees of Wealth or Devotion.” Dasent
(1870): “Everyone had made up their minds. . . .” Mallock, in
ifew Rep. (1878): “Everyone then looked about them silently.”
Jane Austen uniformly employs this usage.
One speech teacher who marked this expression 2 says: “There
is some justification for marking this 1 .” Another says: “Any pro¬
noun here makes me uncomfortable.”
A linguist comments: “The objection to everybody . . . their is
DISCUSSION OF SPECIFIC ITEMS 7S
largely theoretical. The best of writers have used it, and it fills a need.”
A British linguist, in classing this as acceptable for cultivated
colloquial English, says: “But his doesn’t sound pedantic to me, and
I think I say his myself.”
Although a significantly large number of judges approved the
expression, there is sufficient majority against it to indicate that it is
not yet in good standing.
4 . Neither of your reasons are really valid. ( 130 : disputable)
A linguist says, “1 or 2—Older English, but still widely used by
good authors.”
The teachers ranked this lower than did the linguists; in both
groups there was no decided disposition to place this definitely among
illiterate usages.
5 . Neither author nor publisher are subject to censorship.
( 164 : disputable)
This sentence is from Galsworthy’s The Inn of Tranquillity .
Linguists disagree about this expression. Two comments are: “Oc¬
curs in illiterate writing only. Loose colloq.” “Not 1 , but not 2 either,
because it isn’t colloquial enough.”
The Standard Dictionary says: “Grammatical accuracy requires
the use of a singular verb after the pronoun neither; this rule, how¬
ever, is often disregarded in practice and infringements of it may be
cited from good writers.”
The Oxford Dictionary cites Johnson, Cowper, Southey, and Rus-
kin to show the use of neither . . . nor with two singular subjects
and a plural verb. It is curious that, with so much evidence in favor
of this usage from their most distinguished colleagues, the authors
among the judges were almost unanimous in condemning the expres¬
sions as illiterate. In fact, every group of judges placed this lower
than did the linguists.
Apparently this expression is losing ground, and is not in such
good repute as once it was.
IV. Reference
1. I have no prejudices, and that is the cause of my unpopu¬
larity. (21: established)
76 CURRENT ENGLISH USAGE
25 per cent of the judges approved this as formal English; nearly
all the rest approved it as colloquial.
2 . I went immediately into the banquet room, which was, I
found later, a technical error. ( 47 : established)
In this instance the English teachers were less tolerant of the
expression than were the linguists. Both groups, however, unite in
approving this as established colloquial usage.
3 . “You just had a telephone cal!.’* “Did they leave any mes¬
sage? 5 " ( 31 : established)
The great majority of judges approved this as cultivated colloquial
English. See also section on impersonal personsals.
4 . One rarely enjoys one's luncheon when one is tired. ( 4 :
established)
Comments by linguists:
“One followed by one or more one's never wholly excluded one
followed by one or more he's. A series of one's strikes many (includ¬
ing me) as a kind of pedantry. As a matter of fact, probably most
people who stick rigidly to one have acquired it by effort.”
“I rate this 1, but in effect it tends to 4, as semi-literate straining
for correctness.”
The comments and rating on this expression imply that, while cor¬
rect, it is somewhat stilted.
5 . One rarely likes to do as he is told, (in: disputable)
Comments:
“No such phrase used.” {Editor)
“One is the proper form.” ( Linguist)
“Kruisinga points out that this construction is to be found in
books; i. e.—‘You know, my dear, if one esteemed such a person
very much, and were quite sure, without any doubt, that he liked
you in return . . .’ (Meredith, Evan Harrington, Chap. 17). But
generally speaking I should think it is to be discouraged.” {Linguist)
“The English language's lack of a pronoun that satisfactorily com-
DISCUSSION OF SPECIFIC ITEMS 77
bines he, she, and it, to my mind justifies the use of the masculine
he in many cases where it is not quite correct. 55 ( Business man)
The frequent use of such a pronoun sequence as this is probably
due to the desirability of avoiding such stilted locutions as that in
item 4 (One rarely enjoys one’s lunch when one is tired.). Every
group of judges except the linguists regarded such usage as estab¬
lished. Probably it is quite correct, although perhaps the exception¬
ally careful speaker will by rephrasing avoid both stilted correctness
and the looser construction illustrated here.
V. Case
1. It was I that broke the vase, father. (2: established)
“Suggests the speaker who is so afraid of his English that he
pushes ‘correctness 5 beyond the limit. I so frequently hear ‘for you
and I,’ etc., from speakers who have been over-corrected in the matter
of ‘it’s me.’ ” {Linguist)
No judge disapproved this expression; all but 7 rated it I.
2. It is zpe. (73: established)
This is a construction which has been made the subject of news¬
paper editorials beyond counting; and every purist who has felt the
sanctity of grammatical English threatened has gone forth to do
battle against those who would permit the verb to be thus to be
followed by an objective pronoun. The fact seems to be that sche¬
matic grammar has little to do with usage.
Many of the comments recorded were flatly contradictory. Here
are some of them:
“Unpardonable grammar.”
“Incorrect— bad —but used often by discriminating people who
rebel against the formalism of ‘it is I.’ I prefer ‘it is I.’ ”
“Many purists approve it, but it seems not to have gained re¬
spectability.”
“This expression is used so commonly that, among certain classes
of people, it is considered quite correct. Others, however, never
use it.”
“Emerging into 1 (literary English)
78
CURRENT ENGLISH USAGE
“/ sounds quite mad in certain cases; e. g., pointing to a photo:
‘Which is I?M 11 ‘Oh, I see, that’s VI1 ! Absolutely non-English, hang
all the grammarians on earth.”
This expression is listed here among the established usages on
the basis of the way the linguists voted—only three of twenty-eight
condemning it as illiterate. If all the judges’ estimates had been
taken into consideration, without weighting on the basis of the greater
expertness of one group as against another, this sentence would have
been placed among the disputable usages—only the business men,
of whom eighteen condemned and five approved, would place it among
expressions clearly illiterate. One hundred thirty judges altogether
approved this; ninety-one condemned. This can hardly constitute
sufficient reason for taking time to teach “it is I” in school. As a
matter of fact, both forms are at present avoided by careful speakers.
Further light on the importance of formal grammar as a guide to
usage will be gained by considering the grammatically equivalent but
not equally accepted expression it is Mm.
3 . If it had been us, we would admit it. ( 102 : established)
This is a borderline expression, approved by twenty-nine judges;
condemned by nineteen. Certainly there is not enough evidence
against it to make it possible to condemn it dogmatically as bad
English, with such a majority in its favor, although it is not on a
par with “It is me.”
4 . Ill swear that was Mm. ( 138 : disputable)
This also falls among the list of usages that are disputable, but
not unquestionably wrong.
5 . I suppose that’s Mm. ( 152 : disputable)
One linguist says: “Probably less firmly established than that’s
me!’
Another remarks: “I am rather uncertain about the pronoun after
to be. Tt’s me’ I use and consider correct. ‘It’s us’ I have misgivings
about, but I think I have used it or would use it. Tt’s him’ I like least
of all.”
It seems safe to say that grammar has little, if anything, to do
with the validity of such constructions as this. A review of The
DISCUSSION OF SPECIFIC ITEMS
79
judgments on other similar expressions (see items 1, 2, 3, 4 in
this section) will show clearly that “correctness,” in such cases at
least, is entirely a matter of usage, and has little to do with logic and
less with grammar.
This particular expression, while approved as educated colloquial
usage by a significant number of judges, has not won the approval
accorded the objective form of “me” or even “us” after forms of
“to be.”
6. You are older than me. (124: disputable)
The English teachers rated this lower than did the linguists. Of
the latter, two-thirds approve the expression as good colloquial Eng¬
lish.
7. I am older than him. (181: disputable)
Linguists say:
“Personally I generally say T am older than he is! But never
‘older than he! Sometimes, no doubt, ‘older than him.’ ”
“We all know that these expressions are taboo. Also that most
people (educated or otherwise) use them to the exclusion of the
alternate form.”
“4, possibly merging into 2.”
Speech teachers and business men place this expression at the
bottom of the list of expressions on the first ballot; the other groups
of judges place it higher, but there is a decided majority against its
inclusion among allowable expressions.
8. A woman whom I know was my friend spoke next. (194:
illiterate)
The judges’ rating places this expression definitely among un¬
cultivated usages. Cf. Jespersen: Philosophy of Grammar, Appen¬
dix A.
9. I took it to be they. (116: disputable)
A linguist says: “I consider them the ‘correct’ form, but I fancy
people use they a good deal. I don’t consider this form as bad as
‘for she and 1/ etc.”
80
CURRENT ENGLISH USAGE
The linguists rated this expression considerably higher than did
the English teachers, a large majority of whom condemned it as
illiterate. It cannot be considered either as established or as definitely
uncultivated.
10 . I know it to be he. ( 127 : disputable)
A linguist says: “This strikes me as a concocted phrase that would
hardly be used. People say either T know it's he (or him)? ”
A small majority of the judges consider this expression as uncul¬
tivated.
11. It seems to be them. (149: disputable)
An author says: “Stevenson and others use this but it jars me.”
A British linguist says: “I don’t think I should ever say Tt seems
to be they? ” Another comments: “Probably fit seems to be them’
and fit is them’ are not quite in so good use as fit is me.’ ”
A majority of the judges condemn this expression as uncultivated,
although 35 per cent approve it as good colloquial usage. It is obvious
that this expression has not the same standing as “it is me,” which
seems to demonstrate that formal grammar has little to do with cor¬
rectness in matters of this sort.
12. Invite whoever you like to the party. (81: established)
A linguist remarks: “The indefinite who (ever) seems to follow
the same law as the interrogative.”
The linguists rated this expression higher than did the English
teachers, only two out of sixteen condemning it. Thirty-three of the
forty-eight judges approved this use of the nominative case as accept¬
able colloquially.
13. Who are you looking for? (74: established)
The linguists rated this higher than did any of the other groups of
judges; the other groups placed the expression among disputed usages.
All the groups except the business men and authors gave majorities
for approval.
N. E. D.: “Common in colloquial use as object of a verb or preposi¬
tion following at the end of a clause.”
DISCUSSION OF SPECIFIC ITEMS
81
Comments by linguists:
“In our best literature this construction has been waning for cen¬
turies. Whom is now the literary form.”
“Many grammarians fail to note that only the interrogative, not
the relative who is in good colloquial use as objective case.”
“ ‘Whom are you looking fori non-English. ‘For whom are you
looking’ possible book English, but only with some prepositions; viz.,
when the preposition is not felt to make, with the verb, a compound.
I couldn’t say or write ‘without whom can we do?’ I should write, as
I should say, ‘Who can we do without?’ I should write, too, ‘Who are
you looking for?’ etc., like Shakespeare.”
Apparently this is acceptable in informal spoken English, but most
authorities do not approve it for written English.
14. All came except she. (197: illiterate)
A linguist says: “There is a historical basis for the nominative
with both but and except, though for different reasons. Cf. Chaucer,
Cl. T. 508. But probably present-day use is based on an effort to
be ‘correct.’ ”
Of fifteen linguists, two considered this expression allowable; the
English teachers were unanimous in condemning it.
15. Everybody’s else affairs are his concern. (159: dispu¬
table)
Comments by linguists:
“Artificial.”
“Pedantic.”
“Nowhere used.” (British)
“Not English—pseudo-correction by the semi-literate for every¬
body else’s, which is good colloquial English.”
“Curme has here pointed out the true syntactical principle that has
led the popular instinct: the genitive sign immediately precedes the
governing noun.”
It is significant that English teachers, possibly influenced by the
pronouncements of sundry handbooks, would place this among es¬
tablished usages. Over half the linguists, on the other hand, consider
the expression as illiterate or semi-literate. There can be no question,
at any rate, that “everybody else’s” is infinitely to be preferred. See
note on Articles, item 1, in this chapter.
82
CURRENT ENGLISH USAGE
VI. Most as an Adjective
Most anybody can do that. ( 169 : disputable)
All the dictionaries and most of the linguists agree in classifying
this expression as dialectal, although one expert cites a sentence from
Harper's Magazine beginning “Most anyone at the Peace Confer¬
ence . .
The judges were about equally divided between approval and con¬
demnation of this expression.
VII. Much as a Pronoun
This much is certain. (25: established)
This expression was unanimously approved, the majority of judges
rating it as cultivated colloquial English.
Verbs
I. Classes: Transitive and Intransitive Conjused
1. I must go and lay down. (206: illiterate)
One linguist remarks that, while this expression is not now sanc¬
tioned by usage, it was “good in the 18th century.”
There was little disagreement among the judges on this expression,
over 93 per cent of them disapproving it.
2. I had hardly laid down again when the phone rang. (225:
illiterate)
This expression was nearly unanimously condemned as vulgar.
3. The sailors laid out along the yards. (106: established as
technical)
As this is recorded in the dictionaries as a technical nautical term,
it might have been so listed; but it was so ordered by only one out
of five judges. Most of the remaining 80 per cent condemned the
expression, in spite of such dictionary authority as: “Lay, v.i. ... 2.
Naut. to place oneself in a certain position.” (Standard Dictionary.)
83
DISCUSSION OF SPECIFIC ITEMS
4- Just set down and rest awhile. (228: illiterate)
Set, used intransitively in this way, is an undoubted solecism in
present usage.
5 . The neighbors took turns setting up with him. (229: illit¬
erate)
A solecism.
II. Tense
A. Confusion Among Forms of
a. future: shall and will
1. My colleagues and I shall be glad to help you. (46: estab¬
lished)
One linguist says of this expression that it is “not illiterate but
pedantic.” A British linguist says the expression is “not used.”
There is a curious disagreement on this expression between the
English teachers and the linguists. Of the teachers, none disapprove,
and 75 per cent consider it appropriate to the most formal uses. Of
the sixteen linguists, four disapproved altogether, while the remain-
ing twelve were evenly divided between approval as formal literary
English and as good colloquial usage.
Such disagreement among experts, while exhibiting a strong tend¬
ency towards complete approval, gives little justification for dogma¬
tism on the subject of “shall” and “will” by teachers.
2. I will probably come a little late. (135: disputable)
N. E. D.: “In the first person, shall has, from the early Middle
English period, been the normal auxiliary for expressing mere fu¬
turity.”
Comments by linguists:
“Soon will be acceptable as literary.”
“Creeping into current use—disliked by elder people. It was con¬
sidered a Scotticism.”
“This is still ‘popular illiterate’ speech to me, but I believe that
a generation or two will see the distinction lost.”
84
CURRENT ENGLISH USAGE
“Personally, I should say shall, here. But I fancy I am one of a
minority, and I am quite familiar with will.”
“I find few who use shall according to correct rules.”
“Good. Will, I take to refer to one’s own volitions; shall to out¬
side influences.”
These comments show how unwilling experts are to dogmatize on
the distinction (if any) between shall and will. The whole matter is
at present surrounded by a cloud of uncertainty. The only thing
about which there seems to be no doubt is that the hard and fast
rules laid down by most rhetorics and handbooks are not to be relied
on; probably what distinction ever has existed is gradually disap¬
pearing. About two-thirds of the judges approved this particular
sentence.
3 . Yd like to make a correction. ( 53 : established)
A linguist remarks: “I would like is a well established idiom with
those particular about the distinction between shall and will. I have
suspected that the old construction me would like had something to
do with this. I do not agree that there is no historical basis for the
modem distinction, though probably there were different lipes of de¬
velopment in different dialects.”
Only one of forty-seven judges condemned the expression as il¬
literate; nearly all the rest approved it as colloquial.
4 . Will you be at the Browns’ this evening? ( 58 : established)
A linguist says: “2 (cultivated colloquial English) if inviting.
More doubtful if questioning—though shall you sounds affected to
me (not pedantic, just affected, tony).”
This is generally approved as colloquial English. In fact, nearly a
quarter of the judges considered it appropriate for formal use.
b. PAST
1 . He begun to make excuses. ( 200 : illiterate)
The Standard Dictionary says, “Begin; began or begun.”
The N. E. D. says: “Began, established as the standard form; the
alternative begun has also come down to the present day.”
Only 5 per cent of the judges (all from among the linguists, Eng-
DISCUSSION OF SPECIFIC ITEMS
85
lish teachers, and members of the M. L. A.) approve this form even
for colloquial English; it is condemned as illiterate by the remainder.
2. He drunk too much ice water. (195: illiterate)
Dictionaries all record the fact that this form was once correct;
the N. E. D., for instance, says, “Occasional to 19th century.”
Linguists and members of the Modern Language Association, prob¬
ably because of their awareness of the historical justification for this
form, place it higher than do the other groups of judges, although
even they disapprove it by large majorities. Evidently it has almost
altogether lost the standing it once had.
3. Somebody run past just as I opened the door. (227: illiter¬
ate)
The condemnation of this expression was nearly unanimous.
4. She sung very well. (183: disputable)
Comments by linguists:
“I didn’t feel able to characterize sung, etc. I don’t use them my¬
self and they seem to me rather like vulgarisms, but they are his¬
torically as good as sang, etc., they have plenty of good recent lit¬
erary currency, and they may occur colloquially to a greater extent
than I have ever noticed.”
“With the best intentions, I find that my numbering is not con¬
sistent. The preterite drunk I have seen so often that I put it under
2 , but sung and begun I have consigned to A —where, by my own
feeling, all three of them belong.”
“The past tense sung seems to me perhaps slightly better than
begun and drunk, but it is probably used now—except in poetry—
only by old-fashioned people.”
The New International, Standard, and Oxford Dictionaries give
both sang and sung as preterite; the later adds: “Recent usage . . .
has mainly been in favor of sang”
With the exception of the English teachers, who rank this expres¬
sion as nearly established, the judges place this rather low on the
list. Once correct, it seems to be going out of fashion in favor of
“sang.”
86
CURRENT ENGLISH USAGE
5. They swang their partners in the reel. (230: illiterate)
All but one of the judges (all the linguists) rated this expression as
illiterate. See also items 1, 2, 3, 4 in this section.
6. It was dark when he come in. (209: illiterate)
This expression was rated illiterate by a practically unanimous
vote.
7. He dove off the pier. (131: disputable)
The dictionaries characterize dove as colloquial; the N. E. D. adds
that it is “U. S. and Eng. dial.”
A linguist says: “Good colloquially; perhaps acceptable as lit¬
erary.”
Another says: “Dove seems to be regularly used even in good
writing. I fancy dived seems archaic and biblical to most people.”
There was more disagreement among the judges about this than
about most expressions. The general trend, however, seems to be
toward its acceptance, though it is not yet fully admitted to the
category of accepted written usages. #
8. They eat [et] dinner at twelve o'clock. (137: disputable)
Standard Dictionary: “ate or eat (ate is now preferred by many
as the past tense of eat, but the usage is debatable).”
New Int. Dictionary: “pret. eat, ate (in Eng., commonly et)”
N. E. D.: “The pronunciation et is commonly associated with the
written form ate, but perhaps belongs rather to eat, with shortened
vowel after analogy of weak verbs read, lead, etc.”
Comments:
“Would not use this myself.” (Canadian linguist)
“I always say i:t, et, i:tn, never eit.” (British linguist)
“This is good British English.” (British linguist)
“The preterit of to eat is pronounced et in England, but I am
aware that this is vulgar in America. It is generally spelt ate, but
I have an idea it may be spelt eat thus falling into the category of
read, rid, read, red. But perhaps your point is whether it is befitting
to use the verb to eat and the names of meals as the direct object.
Educated usage is, I think, to have dinner, etc.” (British linguist)
DISCUSSION OF SPECIFIC ITEMS 87
“This English form is despised in the Southern U. S.” (Speech
teacher)
“Good English, bad American.” (Author)
This expression is clearly entirely correct in England, incorrect in
the United States.
9 . He looked at me and says . . . ( 205 : illiterate)
96 per cent of all the judges regarded this as uncultivated or il¬
literate.
C . PERFECT
i. John had awoken much earlier than usual. ( 188 : illiterate)
Comments by linguists:
“Sounds grotesque in U. S. Correct in England.”
“Older English, now receding.”
For the United States, at any rate, this expression is not in good
repute.
2 . I have drank all my milk. ( 210 : illiterate)
“Good English in the 17th and 18th centuries,” says a linguist;
“the present use is, I think, not a survival, but a sophistication.”
This expression is not now approved by educated people. See drunk,
sung, begun in section on “Confusion as to Forms of Past Tense.”
3 . They have gotten a new car this year. ( 113 : disputable)
Both linguists and dictionaries testify that this form is acceptable
in the United States, although it is nearly obsolete in England. One
linguist remarks: “The participle form gotten is the usual form in
older English, and naturally the Pilgrim Fathers brought it with them
to New England. It is the usual form in the colonial days and still
the usual American form except in its function as auxiliary, where it
is naturally replaced by got. I have collected a convincing list of ex¬
amples of gotten from our best writers. There is no doubt that gotten
is established in this country. In England it has almost disappeared.”
88
CURRENT ENGLISH USAGE
4 . My contention has been proven many times. ( 109 : dispu¬
table)
Roth dictionaries and judges differ widely on the propriety of
proven as an acceptable variant of proved. The Standard calls it
“archaic”; the New International lists it as a variant of proved; the
N. E. D. says that it is “used properly in passive.”
The linguists, authors, and editors place it among the disputable
usages; the other groups of judges regard it as established. One lin¬
guist says: “Would not use this myself.” Another: “The sound of
proven possibly helps to explain its use over proved. 37 A British lin¬
guist calls it “affected”; a southern linguist remarks that proven is
“general in Arkansas.”
It is hardly possible, in view of this uncertainty, to classify the
word authoritatively, except to say that it cannot be regarded as
illiterate, though “proved” seems to be generally preferred.
B. Sequence of
1 . Galileo discovered that the earth moved . ( 12 : established)
One judge says: “This attraction is too common to be condemned.”
This expression was rated 2 (cultivated colloquial) by a majority
of the judges. One-third of the judges, however, rated this 1 . It is
evidently perfectly correct.
2 . I wouldn’t have said that if I had thought it would have
shocked her. ( 145 : disputable)
The judges were almost evenly divided between approval and con¬
demnation of this expression. It seems to belong in that class of ex¬
pressions which are careless but not absolutely incorrect.
III. Moods: the Subjunctive
1 . If it wasn’t for football, school life would be dull. ( 63 : es¬
tablished)
A great majority of the judges approve this use of the indicative
as good colloquial usage.
DISCUSSION OF SPECIFIC ITEMS 89
2. I wish I was wonderful. (From Barrie’s Dear Brutus )
(93: established)
It is probable that less than fifty years ago the judgment upon
this expression would have been quite different; the decay of the
English inflected subjunctive is vividly illustrated by the fact that
two-thirds of the judges approved this as good informal English.
IV. Agreement with Subject
1. The kind of apples you mean are large and sour. (77: es¬
tablished)
Only one from among sixteen linguists condemned this as illiterate;
the rest considered it colloquially correct. The English teachers con¬
curred, in general.
2. The fire captain with his loyal men were cheered. (166:
disputable)
Comments by linguists:
“There is good literary authority for classing this as 2.”
“Mot 2, but not 2 either, because it isn’t colloquial enough. 5 ’
“Here with is equivalent to and, or analogous semasiology. 55
This expression is certainly ungrammatical, yet in informal con¬
versation would probably be used by educated people. It would prob¬
ably fall under the classification of 2.
There were few expressions on which so much disagreement was
manifested among the judges; out of a hundred expressions, this was
ranked all the way from 34 down to 82. A significant number (13)
of judges (but no linguists) classified this as good literary English,
but there was a small majority in favor of classifying it as unculti¬
vated, not, however, enough to establish it as such.
3. One of my brothers were helping me. (213: illiterate)
This expression was very definitely rated as illiterate.
4. There was a bed, a dresser, and two chairs in the room.
(90: established)
It is noteworthy that the speech teachers ranked this higher than
did any other group of judges. Business men and editors ranked it
90
CURRENT ENGLISH USAGE
lowest. Authors, in most cases the most severe group of judges,
ranked this relatively high.
A linguist writes: “In good use from early times (OE). An inter¬
esting sample is: William Vaughn Moody, Milton’s Complete Poeti¬
cal Works, p. 32: I heard Israel Gollancz say, ‘There was the Chief
Justice and many distinguished men . . .’ ”
This expression cannot be considered wrong in informal cultivated
English speech.
5. It don’t make any difference what you think. (160: dispu¬
table)
This expression is still to be avoided, but it cannot definitely be
placed among the illiterate usages, in face of its approval by nearly
40 per cent of the judges.
6. Martha don’t sew as well as she used to. (174: disputable)
Over one-third of the linguists approved this as colloquial English;
the proportion of English teachers approving it was a little lower.
“Doesn’t” is apparently, by a widening majority, the approved lo¬
cution. *
7. Aren’t (’nt or rnt) I right? (154: disputable)
An American linguist says: “The English seem to have succeeded
in putting over aren’t L I still do not care for it.”
Comments by British linguists:
“Kittenish.”
“British colloquial, coming into use in the U. S.”
“I say this in familiar speech. I shouldn’t write it. I fancy the
majority avoid it.”
“Acceptable as colloquial usage—also ain’t l”
“Ant } no apostrophe (a :nt). I should spell (a :nt) ant without
apostrophe (unless I were writing to a purist, in which case I should
spell it aren’t)’’
The linguists rated this expression considerably higher than did
any of the other groups of judges. This is perhaps influenced by the
number of Britons among the linguists. The expression is evidently
good colloquial usage in England, but has not yet found acceptance
in the United States.
DISCUSSION OF SPECIFIC ITEMS
91
8 . You was mistaken about that, John. ( 220 : illiterate)
It is of interest that none of the expressions on Ballot I received
a unanimous vote of all the judges, either for approval or disapproval.
This expression and “wa’nt” were the only ones that no linguist ap¬
proved.
A linguist says: “Good 100 years ago.”
An author remarks: “In Fielding and others of his time you was
was used of one person; you were of more than one.”
Those who appeal to historical considerations for the defense or
rejection of any expression must be given pause by the fact that this
form, approved by only one per cent of the judges, has an immaculate
historical justification.
V. The Infinitive
a. SPLITTING
1 . The invalid was able partially to raise his body. ( 5 : es¬
tablished)
Linguists say: “This is pedantic”; “I usually split the infinitive in
colloquial speech”; . . But not commendable.”
Correct but stilted.
2 . We can expect the commission to at least protect our in¬
terests. ( 55 : established)
Business men and English teachers ranked this higher than did
the linguists; authors and speech teachers, who alone considered it
disputable usage, ranked it lowest.
One speech teacher, without classifying the sentence, questions it
as ambiguous. Another says: “Not a question of usage but of co¬
herence.”
A business man says: “I have little sympathy with the objection
to the split infinitive. As a matter of fact, I believe a split infinitive
with a word modifying the verb frequently adds strength as well as
clarity to the sentence.”
Comments by various linguists:
“This still connotes illiteracy to me, but I think it is only because
one of my old professors taught me to slander it.”
“Many writers split infinitives at will, and defend the practice.”
92 CURRENT ENGLISH USAGE
“The particular sentence is awkward—but a split infinitive is
O. K., as such.”
“In this particular phrase the split infinitive seems to me rather
unhappy. Sometimes it is perfectly legitimate and even necessary.”
“I do not like the cleft infinitive but it is infinitely used.”
“The use of the split infinitive is in fact sufficiently common in
good writing to class it in 1. Common sense suggests its avoidance
when nothing is gained in clearness.”
“ ‘So as to always fit 7 (Dean Swift). Swift’s occasional splits are
not a conclusive argument. But the thing to note is that Swift, fol¬
lowing his instinct for style, saw that it was better to split ‘to fit’
than to split ‘so as to. 7 Not so your purists. They will give you ‘owing,
however, to 7 because no rule of thumb forbids. So they see no ob¬
jection to ‘used often to go 7 which to me is non-English, whereas
‘used to often go 7 only offends against the rules of the porridge¬
brained (i. e., grammarians) and is English. Further, the purists
never even notice the split in ‘he did it even though he oughtn’t to 7
—(they used to, but now they don’t)—nor in ‘to come and go. 7 77
The evidence in favor of the judiciously split infinitive is suffi¬
ciently clear to make it obvious that teachers who condemg it arbi¬
trarily are wasting their time and that of their pupils.
b, DEBATED PHRASES
1. We will try and get it. (49: established)
Except for the speech teachers, a majority of whom condemned
this expression, the judges for the most part approved this as ap¬
propriate for cultivated colloquial English—10 per cent of them con¬
sidered it as belonging to literary English. Milton employs it. All
dictionaries sanction it as colloquial.
A British linguist says: “Try and but not tried and, tries and, etc.
Only the form try —e. g., will try and —an important observation
purists miss of course.”
This expression is evidently perfectly correct for cultivated collo¬
quial use.
2. I want for you to come at once. (216: illiterate)
Nineteen out of twenty judges in all groups rated this expression
as illiterate. It is, however, in cultivated use in the South of the
DISCUSSION OF SPECIFIC ITEMS 93
United States. This would seem to indicate that, in the estimation
of the judges, dialect usages are generally equivalent to illiterate.
3. The kitten mews whenever it wants in. (193: illiterate)
Linguists and dictionaries agree that this expression is dialectal in
certain localities—chiefly Scotland and here and there in the United
States. In those localities where it is current, it may be allowable as
a popular colloquialism; it has no standing for more formal or gen¬
eral use.
VI. Participlesj Debatable Uses of
1. I will go, providing you keep away. (86: established)
Although 135 judges approve this, as against 83 who disapprove,
all the groups except the linguists place it among the disputable us¬
ages.
A British linguist remarks: “To me this sounds vulgar but is
gaining ground in newspapers (which generally, I may mention, set a
pretty high standard in England) and is given in Concise Oxford
Diet, (vide provide, ad. fin. ‘providing that’—foil.). The following
word is provided; the Cone. Oxf. seems therefore to suggest that
provided is preferable, though it does not, as I should, condemn pro¬
viding . 39
The N. E. D. says: “providing, present participle. Quasi-conjunc¬
tion (without that). On condition that; in case that; if only.”
With so decided a majority in its favor, and with such dictionary
evidence to support it, it would be hard to justify any campaign to
eliminate this expression from the vocabulary of school children.
2. The child was weak, due to improper feeding. (187: dis¬
putable)
The linguists and the members of the Modern Language Associa¬
tion voted about two to one against the inclusion of this expression
among the approved usages; the other groups of judges gave a con¬
siderable majority in its favor. There was wide disagreement among
the groups as to the proper placement of the expression; it must be
included, therefore, among the disputed usages.
94
CURRENT ENGLISH USAGE
Linguists remark:
“Good colloquial English in the sense used.”
“I dislike this, but it is certainly in wide use.”
“'Used by all military writers.”
“This would be unobjectionable in England, but I imagine that in
America it would sound much as T guess’ does to us. I think the
American usage is better.”
“Due to is particularly annoying to me; but it is rapidly gaining
headway.”
The N. E. D. lists this expression as “rare before the 19 th cen¬
tury.”
VII. Gerund, Possessive with
i. What was the reason for Bennett making that disturb¬
ance? (95: established)
A linguist says: “This is difficult to classify because it is con¬
sidered good usage in British English, but the possessive seems to be
used with proper names in American English.” [Perhaps for reasons
of euphony.]
Another linguist comments at some length: “Tor me # making’
sounds wrong, and yet not vulgar. A kind of careless colloquialism,
rather than a solecism. Tor it making’ sounds correct, but I think
I say Tor its making.’ With nouns I fancy I say Through the dog
making such a noise’ as often as ‘dog’s.’ I would never correct a child
for saying ‘dog,’ especially as it is more logical. It is not the making
which happened to be the dog’s that’s in question, but the dog-making
fact. Making cannot be mistaken for a participle, because in that
sense we should say ‘which was making.’ Even if it were mistaken for
a participle, it is a less evil than the other word (not word’s please)
being mistaken for a plural. Moreover, ‘dog making’ allows an im¬
portant distinction:
( 1 ) ‘I was surprised at Bennett bowling’ (that he bowled).
(2) ‘I was surprised at Bennett’s bowling’ (that it was so good
or bad).”
Each group of judges except the authors gives a decided majority
for approval of this expression as good colloquial English.
DISCUSSION OF SPECIFIC ITEMS 95
2. What are the chances of them being found out? (125: dis¬
putable)
Comments, by linguists:
“Illiterate, but some seemingly parallel cases are not objectionable.”
“Not the best use, but defensible.”
“Personally, I should use the full gerund (their being) here. But
this construction doesn’t shock me, and is, I think, very common.”
This expression can hardly be considered apart from the one above.
(“What was the reason for Bennett making that disturbance?”) It is
interesting to note that in the latter instance the linguists ranked
the expression lower than did the other groups of judges, although
they approved it by a large majority; whereas the linguists ranked
the expression now under consideration much higher (17 for, 10
against) than did any of the other groups. Sentences 1, 2 and 3 in
this section are grammatically similar; yet 2 and 3 are decidedly less
approved than 1.
Apparently the possessive of a proper name before a gerund is less
obligatory than that of a pronoun. The above pronoun form, while
passable for the most informal English, is to be avoided.
3. That was the reason for me leaving school. (136: disputa¬
ble)
A linguist says that this is entirely correct, “but not commenda¬
ble.”
Opinion is nearly evenly divided as to the standing of this ex¬
pression. In the present state of usage, it cannot definitely be said
to be either decidedly right or wholly wrong.
VIII. Faulty Verb Forms
1. I suppose I’m wrong, ain't I? (172: disputable)
A British linguist says: “Good colloquial English, but old fash¬
ioned.”
Six of seventeen linguists considered this expression appropriate
to the colloquial uses of educated people; the English teachers were
nearly unanimous in condemning it. It stands very near the line
established in this study between the disputable and the entirely dis¬
reputable usages. See also aren't 1, etc.
96 CURRENT ENGLISH USAGE
2 . Ain't that just like a man? ( 207 : illiterate)
There is clearly a distinction in usage between this expression and
ain't with the first person singular, where a commonly accepted in¬
terrogative contraction is lacking. As used here, “ain’t” is almost
unanimously condemned.
3. That ain't so. (203: illiterate)
Nearly all the judges condemned this form.
4. My cold wa’nt any better next day. (223: illiterate)
A British linguist says: “Purely American to me.”
An American linguist remarks: “This is used by some northern
New Yorkers, as I have heard; I have heard it used by a few in
North Carolina; I think it is decidedly illiterate.”
Only two among over 200 judges approved this for colloquial use.
Decidedly it is not justified for use by educated people.
5. The stock market collapse left me busted, (163: disputa¬
ble)
Some support for this expression as colloquial English may be
found among the linguists; the English teachers are almost unani¬
mous in condemnation.
6. The dessert was made with whip cream. (191: illiterate)
75 per cent of all the judges rated this expression as illiterate.
IX. Nouns Made Into Verbs
1. We taxied to the station to catch the train. (65: estab¬
lished)
The linguists were unanimous, and the English teachers nearly so,
in classifying this as cultivated colloquial English.
2. He stopped to price some flowers. (70: established)
Two-thirds of all the judges regard this expression as acceptable
colloquial English.
DISCUSSION OF SPECIFIC ITEMS
97
3. He loaned me his skates. (99: established)
The British linguists remark that the verb “loan” is not used in
England. Forty-one of the forty-seven judges, however, approve the
expression as acceptable.
4. The banker loaned me $200 at 6%. (104: established as
technical)
195 of the 229 judges approved this use of loan; a quarter of these
considered it acceptable as formal or literary English. Fifty-four
judges classified it as technical. Their view is supported by the dic¬
tionaries (so long as the term is used in connection with finance);
the dictionaries point out, however, that the use of loan as a verb as
a verb is restricted chiefly to the United States.
The distinction between “loan” and “lend” among bankers and
business men has virtually disappeared in the United States; either
is correct for business uses. The use of “loan” as a verb in other
connections, however, is less approved, although it cannot be con¬
demned as illiterate.
Adjectives
I. Pronominal
1. I was attacked by one of those huge police dogs. (32: es¬
tablished)
The votes of the judges clearly place this among acceptable collo¬
quialisms.
2. Harry was a little shaver about this tall. (83: established)
Over 75 per cent of the judges classified this expression as culti¬
vated colloquial English.
3. Don’t get these kind of gloves. (167: disputable)
The linguists ranked this higher than did any other group of judges.
The editors placed it, by unanimous consent, at the very bottom of
the lfst of usages; the English and speech teachers rated it nearly
as low. Evidently this expression is not at present acceptable as culti¬
vated English in the United States. See also Verbs IV, item 1.
98
CURRENT ENGLISH USAGE
II. Faulty Forms
A light complected girl passed. (215: illiterate)
This expression was almost unanimously condemned.
Articles
1. A Tale of Two Cities is an historical novel. (1: estab¬
lished)
Two or three judges remarked that this expression is archaic or
old-fashioned. It is distinctly literary, as opposed to colloquial—out
of forty-six judges, only ten rated it lower than l. 4 See also a orange.
2. There was a orange in the dish. (218 : illiterate)
This was one of the few expressions on either ballot on which the
vote of the judges was unanimous. There was not one vote for in¬
cluding this form among usages in any way allowable. Contrast an
historical novel, above.
Adverbs
I. Double Negatives
1. We haven't but a few left. (143: disputable)
Here again the English teachers are more severe than the linguists.
The locution is not, however, generally approved by either group.
2. I can't help but eat it. (153: disputable)
Comments by linguists:
“I mark this 4, but I suppose it may be 1; I could not feel sure
without actual search.”
“Very common in England and America and grammatically
formed.”
“I have been studying for forty years the use of help in elliptical
construction in the sense of avoid. T couldn't help (do anything) but
laugh.' Help in this meaning is common apart from its use wij^h the
* This is but one of a number of expressions among the “established usages”
which might be called hyper-urbanisms —artificial, trite, pedantic, or stilted at¬
tempts at correctness.
DISCUSSION OF SPECIFIC ITEMS 99
infinitive: ‘I can’t help it . 5 The Oxford Dictionary recognizes the
construction with the infinitive as grammatical, but remarks that the
infinitive is rare, now being replaced by the gerund. The editor was
not wide awake when he made the remark. The gerund is the usual
form after help, but when the conjunction but is used the old infini¬
tive construction is very common in the best writers of England and
America. I have a large collection of examples from good authors.
I shall publish these materials some day.”
The Oxford Dictionary cites Hall Caine: “She could not help but
plague the lad.” The New International and Century Dictionaries
definitely accept the expression.
Only the authors placed this expression as low as did the linguists,
who were evenly divided between approval and condemnation. Ap¬
parently many cultivated people still have a preference for the gerund
construction here, but the expression as given cannot be called defi¬
nitely wrong.
3 . I havezrt hardly any money. ( 189 : illiterate)
A linguist who condemns this says: “But T haven’t any money,
hardly,’ would be colloquially acceptable.”
Another, also disapproving, says: “But a reguliar idiom in my
speech!”
A British linguist remarks: “Sets my teeth on edge, like ‘Ere he
had scarcely begun . . (Burroughs, Tarzan oj the Apes); ‘Hardly
had he finished, than . . .’ (Sir Edmund Gosse, Father and Son,
incredible though it may appear).”
With very little disagreement among the various groups of judges,
this expression is disapproved for use by educated people.
II. Adjectives Used as Adverbs
1 . Our catch was pretty good. ( 43 : established)
Of forty-eight judges, only one listed this expression as illiterate.
Most of the rest placed it among cultivated colloquialisms.
2 . That’s a dangerous curve; you’d better go slow, ( 56 : es¬
tablished)
15 per cent of all the judges approved this as formal literary Eng¬
lish; the same proportion condemned it as illiterate (but only 4 of
100
CURRENT ENGLISH USAGE
27 linguists); the rest approved it as colloquial. The Standard, New
Int., and N. E. D. all approve slow as an adverb.
3 . Drive slow down that hill! ( 82 : established)
Comments, by judges:
“Might be heard in an excited moment—would hardly be written
without coming under class 4.”
“It is probably being driven at, but slowly.”
“When referring to auto driving the -ly is almost universally
dropped.”
A large majority of judges approved this as good colloquial English,
except among the business men and authors. The reason for its being
ranged lower than “you’d better go slow” (above) is perhaps best
stated in this comment by one of the linguists: “I have marked this
4, though slow of course is used properly enough as an adverb—as
go slow is a commonly accepted informal expression. Followed by a
down or an up, however, I believe one must use slowly. 77
4. My father walked very slow down the street. (89: estab¬
lished)
Three of the linguists classify this expression as illiterate. One of
them says: “But go slow or drive slow is good English.” Another
remarks: “Walk slow is good English; followed by a phrase, as here,
it is uncultivated.”
A linguist who approved the expressions says: “Slow and slowly
give different senses.”
Among these distinctions, apparently based squarely in considera¬
tions of euphony, the fact emerges that “slow” is safely established
as an adverb.
5 . He moves mighty quick on a tennis court. ( 69 : established)
Though the English teachers rate this considerably lower than do
the linguists, a large majority of both groups of judges approve the
use of “quick” as an adverb in this context as cultivated colloquial
English.
6 . Will you go? Sure. ( 133 : disputable)
No judge approved this as formal literary English, but a majority
DISCUSSION OF SPECIFIC ITEMS
101
of both groups considered the expression as valid in the informal
conversation of educated people. It is clearly better independently
used than directly modifying a verb or adverb. See next item.
7. It sure was good to see Uncle Charles. (176: disputable)
A British authority says that this expression is “impossible in
England. 55 Another linguist, who condemns the form, adds, “but in
Milton 5 s prose once upon a time! 55
“Sure 55 has not gained approval as a directly modifying adverb.
8. John didn’t do so bad this time. (157: disputable)
50 per cent of the linguists, and over 35 per cent of the teachers,
considered the expression good usage in colloquial English.
g. Yes, our plan worked just fine. (165: disputable)
A British linguist says: “Not English except to imitate Americans.”
The N. E. D. calls this adverbial use of fine “obsolete except dia¬
lectal 55 and gives 1890 as the date of the latest recorded usage.
A small majority of linguists condemned this expression as un¬
cultivated ; a large majority of English teachers considered it good
as informal English for educated people.
10. I£ I asked him, he would likely refuse. (156: disputable)
Linguists and dictionaries agree that this expression is American
or Scottish, as opposed to British, and it is probably acceptable col¬
loquially in those countries. Only the editors placed this expression
lower than did the linguists; the other judges assigned it a position
near the established usages.
11. It’s real cold today. (178: disputable)
Comments by judges:
“I have found this provincially on well-bred tongues.”
“A little playful. 55
“Is heard in England, but more colonial or U. S. 55
Standard Dictionary: “Colloq. U. S.; an erroneous use.”
New English Dictionary: “Loosely in later use (chiefly U. S. or
Scot). 55
102
CURRENT ENGLISH USAGE
Apparently “really,” “very,” or “extremely” are more acceptable
than “real” in expressions like this.
12. The engine was hitting good this morning. (190: illiter¬
ate)
A British linguist says: “Impossible in England.”
The judges’ vote was over five to one for placing this expression
among illiterate usages.
13. He did noble . (226: illiterate)
The judges unanimously consigned this expression to the class of
illiterate usages. But see also bad, fine, good, quick, real, slow, sure.
Note that these monosyllabic adjectives, except good, fall in the
disputable or accepted usage categories.
III. Misuse of Very
The man was very amused. (147: disputable)
“The borderline between the fully naturalized adjective * (interest¬
ing, amusing, tired, surprised, etc.), and the participle is very difficult
to determine.” (British linguist)
“The use of very with past participles has become more common,
but very amused would not be used by most good speakers.” (British
linguist)
“I do not like very amused . There seems to be a touch of shadowy
elegance about that which can be justified no more than the carrying
of a stick or the wearing of spats.” (British linguist)
“I don’t like this but it seems to be good British English.” (British
linguist)
“Not used. Add much and it might go as colloquial.” (Editor)
“I have seen very plus the past participle in letters of a highly
educated English university man (and frequently elsewhere in Eng¬
land since)(Linguist)
“Spreading rapidly. Much amused sounds pompous—or facetious.
So I say 'very much amused/—I think. (Perhaps I say 'very amused’
without knowing it).” (British linguist) ^
This is an instance where experts disagree. Nearly 10 per cent of
the judges voted that the expression is good literary usage; about
DISCUSSION OF SPECIFIC ITEMS
103
30 per cent considered it acceptable as colloquial; the remainder
condemned it. Probably careful speakers will avoid it, but it cannot
be called a solecism, nor can its extirpation be made a basic element
of school courses in English.
IV. Doubtful Adverbs
1. He is kind of silly, I think. (134: disputable)
The dictionaries all list this expression as colloquial.
Comments by linguists:
“Indispensable.”
“Bad—but ‘sort of silly’ would be acceptable, and perfectly logical.”
“I recognize that kind of has come into bad repute, but this is be¬
cause our grammarians have been men unacquainted with the history
of English. For many centuries kind of has been an adjective element
and is still widely felt as an adjective. It ought to be rescued from
the false feeling that has become associated with it from reading our
English grammars. This consideration weighs with me! No literary
substitute takes its place. We always feel the literary substitute as
inadequate. Then, what shall we do with what kind of if we condemn
kind of?* Everybody who speaks English says ‘What kind of trees
are those?’ Then, the question of the adverb kind of arises. We
surely need it. Kind of used as an adjective led to the use of kind of
as an adverb, just as in general an adjective can be used as an adverb.”
The judges were nearly equally divided as to the classification of
this expression. While it cannot be dismissed as illiterate, it probably
is not altogether a safe usage for cultivated speech. Many will prefer
somewhat.
2. I felt badly about his death, (79: established)
A few of the linguists condemned this as “pedantic”; others ap¬
proved the expression as appropriate to formal, literary English; the
majority of all the judges classified it as an acceptable colloquialism.
3. He most always does what his wife tells him. (175: dispu¬
table)
Here the English teachers were more lenient than the linguists;
the judgment of both, however, would place this expression low among
disputable usages. See also Pronouns VI.
104 CURRENT ENGLISH USAGE
4. My experience on the farm helped me some, of course. (177:
disputable)
Dictionaries classify this expression as U. S. (and probably provin¬
cial English) dialect. A small majority of the judges would approve
it as colloquial; its position is dubious.
5. Well, that's going some . (139: disputable)
British linguists agree that this expression is not heard in England.
Linguists are about evenly divided on this expression; English
teachers do not regard it so leniently. It cannot be said to be ac¬
cepted.
V. Position of Adverb
1. We only had one left. (66: established)
Palmer’s Grammar of Spoken English , 386 , p. 184 , states: “Only
is commonly used in the pre-verbal position.”
One of the disapproving linguists says: “The best English writers
seem to go out of their way to misplace only”
Another linguist comments: “Here I think a difference should be
made in writing and conversation, since the tone of the voice always
indicates in conversation what is limited by only. In this particular
phrase, c we only had one left,’ ambiguity is hardly possible. In many
cases only, even in the best literary style, need not precede what is
modifies. Compare Arnold’s ‘Dover Beach’:
And now I only hear
Its melancholy, long, withdrawing roar.
The propriety of the position of only depends entirely upon the phrase
in which it is used.”
Only two of twenty-seven linguists rated this illiterate. A number of
authors, editors, and business men rated it disputable, although in
the two groups last named a majority approved it. Apparently in¬
struction should attack ambiguous cases only.
2. Cities and villages are being stripped of all they contain not
only, but often of their very inhabitants. (158: disputable)
This is exactly the same construction as that below: Woodrow
DISCUSSION OF SPECIFIC ITEMS 105
Wilson’s use of it probably influenced a quarter of the judges to ap¬
prove it as literary English; a majority, however, disapproved, proba¬
bly owing to the extreme awkwardness of the sentence. For the latter
reason, this is scarcely a test case.
3. His presence was valueless not only, but a hindrance as
well. (179: disputable)
Comments by linguists:
“A Wilsonian, not an English locution.”
“Whether Wilson’s use of this expression should be called a rhetori¬
cal device or an affectation of peculiarity I don’t know; its use by
others I should set down purely to affectation.”
“Not idiomatic, but not incorrect, I suppose, and certainly not
colloquial.”
“I have never heard the expression.”
“Rather a misarrangement than a solecism.”
“I cannot imagine anyone putting not only in this position. If put
before valueless I do not object.”
This expression caused considerable confusion among the judges
because j.t seemed not to belong to any of the suggested categories.
Twenty judges classed it as 1, thirty-one as 2, and about 150 as 4 .
Apparently Woodrow Wilson’s repeated use of it has not established
it as cultivated English.
Comparison
I. Further and Farther
1. I felt I could walk no further . (41: established)
Only the business men and speech teachers place this among dis¬
putable usages. The other five groups of judges consider it as es¬
tablished, their rankings ranging from 12 to 28 .
All American dictionaries give farther and further as synonyms.
The N. E. D. says: “In standard English the word farther is usually
preferred where the word is intended to be the comparative of far,
while further is used where the notion of far is altogether absent;
there? 1 is a large intermediate class of instances in which the choice
between the two forms is arbitrary.”
Comments by British linguists:
106
CURRENT ENGLISH USAGE
“The effort to make a distinction between farther and further is
still unsuccessful.”
“ T could walk no further’ seems quite wrong to me, but not vulgar,
simply wrong.”
“ ‘He went further than I’ (went on beyond the point I went to)—
‘He went farther’ (had a longer walk)—seems quite a natural dis¬
tinction to me, though it was probably pedantry with my father, as
I don’t seem to hear it from other people.”
Apparently if there is any distinction between “farther” and
“further,” it is still too subtle for even experts to be sure of what
it is.
2. This is all the further I can read. (202: illiterate)
Dictionaries do not recognize this expression, and most of the
judges rated it as uncultivated.
II. Superlative Used for Comparative
Of two disputants, the warmest is generally in the wrong.
(141: disputable)
(This is the title of one of Lamb’s “Popular Fallacies.”)
Comments, by linguists:
“Illiterate; though I would differentiate by the comparative in my
own discourse, whereas here the duality is definitely affirmed in the
speech.”
“The use of the superlative of two I find quite generally in the
conversation of British novels. Evidently in England the rhetoricians
haven’t been able to frighten people into avoiding it.”
“This use of warm not natural to me. Ordinary colloquial use of
superlative.”
Apparently Lamb’s (and other authors’) use of expressions similar
to this has not rendered it entirely acceptable. The editors are almost
unanimous in condemning it; the other groups of judges, while not so
severe as the editors, give a majority for classing the expression as
uncultivated, though many approve it as colloquial, and there is even
a scattering of votes for its approval as literary English (possibly by
those aware of its origin).
DISCUSSION OF SPECIFIC ITEMS
107
III. Worse and Worst
In the collision with a Packard, our car naturally got the
worse of it. (144: disputable)
One of the linguists who condemns this as uncultivated says:
“Semi-literate care for logic, with no sense of idiom.”
There was a great deal of disagreement among the judges on this
expression. 20 per cent of the linguists and 30 per cent of the teachers
approved it as formal English, but a majority of both groups con¬
demned it as illiterate.
Prepositions
I. Unusual Uses of
1. A treaty was concluded between the four powers. (75: es¬
tablished)
Only the English teachers ranked this higher than did the linguists.
Thirty-two judges approved this as formal literary English; eighty
approved it as good colloquial English.
In the New Int. Diet., this very sentence (with three powers, in¬
stead of four) is given as an example of the proper use of between
bringing two or more objects severally and individually into the
relation expressed.
N. E. D.: “In all senses, between has been, from its earliest ap¬
pearance, extended to more than two. . . . It is still the only word
available to express the relation of a thing to many surrounding
things severally and individually, among expressing a relation to
them collectively and vaguely: we should not say ... ‘A treaty
among three powers’ . . .”
All the evidence available seems to indicate that the use of “be¬
tween” in such a context as this is perfectly proper. (See also the
remarks on the next phrase, “between each bed.”)
2. There is a row of beds with a curtain between each bed.
(|55 : disputable)
Facetious comment by a British linguist: “This is one of those
expressions even the careless probably feel uncomfortable about. I
108
CURRENT ENGLISH USAGE
have heard ‘between each bed and the next/ ‘between every pair of
beds’ (objectionable, suggests beds in pairs), ‘between every two
beds.’ I use ‘and the next’ because it enables me to say it wrong, and
then put it right.”
Another linguist remarks: “The only sensible locution.”
Authors and editors condemn this expression as illiterate; the
other groups of judges are about equally divided as to its reputability.
Evidently it cannot be definitely assigned either to the established
or to the uncultivated usages.
3. He came around four o’clock. (101: established)
In America, this expression is good colloquial English. In England
it is rarely heard.
4. Under these circumstances I will concede the point. (20:
established)
The standard deviation of ranking among the groups of judges for
this item was only 2 . 62 , and only sixteen of more than 200 placed
it in class 4 .
The expression is evidently perfectly correct.
5. The old poodle was to no sense agreeable. (2x2: illiterate)
This expression, which is from Arnold Bennett’s Old Wives’ Tale , 5
was thought by the compilers of the ballot to be a fresh and meaning¬
ful way of putting the idea. Nevertheless, the English teachers would
have placed it among the disputable usages, and only one of the
linguists approved it even for colloquial English.
II. Much Disputed Phrases
1. As regards the League, let me say . . . (30: established)
The linguists, the speech experts, and the M. L. A. judges con¬
sidered this as established; the other groups of judges rated it as
5 The sentence in The Old Wives* Tale is: “Fossette was to no sense a pleasant
object” It refers to a sick old dog who smelled evilly and was unpleasant to all
the senses. Both the compilers of the ballot and the judges (lacking the context)
have entirely misread this item, supposing the to no sense to mean in no sense .
The vote is thus valueless.—R. M. W.
109
DISCUSSION OF SPECIFIC ITEMS
disputed. The linguists ranked it 9 ; the lowest ranking assigned was
that of the business men (47).
2. Sitting in back of John, he said, “Now guess what I have.”
(115: disputable)
A British linguist says that this expression is “never used.”
This expression cannot definitely be said to be incorrect, but it is
not established as a good usage. See also III, 1 in this section.
3. He stood in front of the class to speak. (24: established)
With only one exception, all the judges approved this expression
as entirely correct. See also III, 1 in this section.
III. Omitted Prepositions
1. The catcher stands back of the home plate. (45: estab¬
lished)
The following comment, by one of the linguists, illustrates the
metaphysical nature of the controversy that sometimes arises over
this expression:
“I maintain . . . that there is a distinction of meaning between
back of and behind. It is not invariable, but they are not always
synonyms. A tree growing in front of a house might be hidden behind
the house to one in the back yard; but it would not then be back of
the house. Is back of always opposite to in front of, whereas behind
is beyond something from the observer?”
More of the judges approved this as formally correct than con¬
demned it altogether; but the great majority placed it among the
cultivated colloquialisms. See also in front of and in back of.
2. He doesn’t do it the way I do. (39: established)
This expression is clearly good colloquial English, but roundabout.
3. J,tme was home all last week. (52: established)
The judges were nearly unanimous in classifying this as an ac¬
ceptable colloquialism.
110 CURRENT ENGLISH USAGE
4. He never works evenings or Sundays. (112: disputable)
Comments, by linguists:
“Rather old-fashioned.”
“In a Yorkshire manufacturing town this might be heard more
generally than in London.”
“May be heard in England, but I suppose ordinarily U. S.”
Nearly three-quarters of the judges approved this expression, and
it is probably quite correct in the United States.
5. Sam, who was then in town, was with me the three or four
Erst days. (Quoted from Lamb’s “Popular Fallacies”) (no:
disputable)
Although a majority of judges approved this expression, it cannot
be placed among the established usages.
IV. Elliptical Constructions With
I enjoy wandering among a library. (214: illiterate)
This sentence is from De Quincey’s “Essay on Style.” It is plainly
elliptical yet only two English teachers rated it as correct for formal
literary use; one linguist approved it for colloquial use; the remainder
of the judges* (93 per cent) condemned it as uncultivated.
V. Redundant Use of
1. We cannot discover from whence this rumor emanates. (50:
established)
A linguist, who classes this expression as colloquially acceptable,
says: “The from seems more redundant than ever in the indirect
question. From whence in other cases might be 1.”
English teachers rate this expression rather low; they are almost
equally divided between approval and condemnation. The linguists
are more liberal, but rather scattered—five marked this 1; seven
marked it 2; while four marked it 4 . One was uncertain whether it
should be 2 or 4 . *
In spite of considerable uncertainty, it seems that this may safely
be classed among established usages.
DISCUSSION OF SPECIFIC ITEMS
111
2. Now just where are we at? (192: illiterate)
One linguist says: “Acceptable as jocose.”
The linguists and English teachers place this lower than do other
groups of judges; no group, however, places it higher than rank 74
out of 100 expressions on the ballot.
3. She leaped off of the moving car. (182: disputable)
A linguist says: “Not in my vocabulary, but it would be saying
too much to call it illiterate.”
The New International Dictionary says, “Formerly in good use,
and in dial, and vulgar use still.”
Not quite one-third of the judges approve this expression as col¬
loquial English. The remainder consider it uncultivated.
Conjunctions and Conjunctive Adverbs
I. Disputed Uses of
1. This hat is not so large as mine. (13: established)
The ratings assigned this expression show that it is entirely cor¬
rect, but see also not as, below.
2. He did not do as well as we expected. (26: established)
Only the M. L. A. judges placed this among the disputed usages
(rank 49 ), but the standard deviation of ranks was over 14 , showing
considerable uncertainty. The mean rank was 18 .
One editor says: “I have marked this sentence 2 (colloquial), for
the reason that the distinction between the use of as and so in positive
and negative expressions simply is not made, though recommended
by careful writers.”
An author says: “So should be used with the negative but some¬
times is awkward.”
The use of as in this construction is established in cultivated Eng¬
lish.
3. Tfyis was the reason why he went home. (34: established)
r
One linguist says: “ £ The reason why’ is all right, even in 1, but
‘the reason was because/ though spreading in newspapers, is bad,
decidedly.”
112 CURRENT ENGLISH USAGE
All the groups of judges regard “reason why 57 as established, and
the great majority place it in class 2 or among colloquial expressions.
4. The real reason he failed was because he tried to do too
much. (80: established)
The opinions of the judges in this instance leave little doubt that
the expression is acceptable colloquially.
5. I don’t know if I can. (60: established)
Thirty-one from among forty-eight judges approve this as collo¬
quial. The remainder are divided between approval as literary English
and condemnation as illiterate. The latter demand whether in this
sentence.
6. Either of these three roads is good. (123: disputable)
Standard Dictionary: “Either is sometimes used loosely for any,
referring to a larger number than two. 77
N. E. D.: “Sometimes equals each (of more than two things). 77
Comments: ,
“This does not sound illiterate to me, but always looks so! 77
(Linguist)
“I should say this, but I doubt if I should commit it to paper, even
in an intimate letter! 77 (Linguist)
“If we had a satisfactory substitute for the word either when we
speak of a choice of more than two things, I should not classify this
as colloquially acceptable. Of course, the word any is satisfactory
except that it has so many uses, whereas the word either implies choice
and for that reason seems to me to be permissible even when this
choice is between more than two. 77 (Business man)
“Either of more than two does not seem to be in sufficient use to
warrant its inclusion as literary English. Yet it does not specifically
belong to any of the other groups. Any or any one seems to serve
well enough. Yet many of the best writers have used either in this
sense. 77 (Linguist)
Although eighteen of twenty-nine linguists approved this depres¬
sion, among the other judges there was a majority who condemned
it as uncultivated. Although it is not definitely among expressions
proscribed by usage, it should probably be avoided on the principle
DISCUSSION OF SPECIFIC ITEMS
113
that where usage is divided, one is more comfortable on the con¬
servative side. This principle, however, probably does not justify drill
to establish the conservative form.
II. Omission in Double Construction
He could write as well or better than I. (119: disputable)
A linguist, in criticizing the categories as indicated on the ballot
(see p. 67 , above), says: “. . . It seems to me that there are a
number of usages which are not in 2 , and yet should not be put in 4 .
I mean usages that arises inadvertently when there is some complexity
or difficulty, or lack of forethought in speaking. I cite the above case.
The speaker runs on and adds better to as well and then is in dif¬
ficulties. If he is writing he can change the order, but in speaking he
must make the best of it.”
Other comments:
“The second as is understood. There is a double comparison and
I believe the omission of as and the use of the word than in such a
case is justified. I have, therefore, classed it as 2 .” (Business man)
“Our language lacks some needed particles for correct conversation
—‘as well as I, even better than F is correct but cumbersome.”
(Linguist)
The fact that the majority of judges approved this expression
makes it impossible to say that it is illiterate; the consensus of
opinion, however, seems to be that it is awkward, and to be avoided.
III. Preposition Used as Conjunction (like and as)
1. We don’t often see sunsets like they have in the tropics.
(180: disputable)
A British linguist remarks: “I think I say ‘like the ones they have 5,
(I am a schoolmaster). But the other doesn’t sound vulgar to me.”
A decided majority of judges condemned this expression as unculti¬
vated, although there were enough votes for acceptance to prevent
its being placed among indisputably illiterate usages.
2. It'looked like they meant business. (185: disputable)
A linguist remarks: “The popular instinct in this and analogous
uses of like is sound; it is more distinctive and clearer than as.”
114
CURRENT ENGLISH USAGE
This expression found acceptance among only a quarter of the
judges. Although not so definitely reprehensible as the usage in the
section headed “uncultivated usages,” it is probably incorrect.
3. Do it like he tells you. (186: disputable)
A British linguist says: “I rate this as good colloquial English—
good literary English where clause-verb is suppressed; e. g. ‘Roared
like a bull. 3 Where like means definitely £ in the very manner 3 1 should
rather say £ Do it the way he tells you, 3 or even £ Do it how he tells
you, 3 though I feel the latter (not the former) to be doubtful—chil¬
dren’s English.
“When I use like it is rather, so to speak, appositional. T ran away
of course, like you did 3 (the same thing which you did).
“When the clause-verb is omitted, everyone uses like (even the
blithering purists—not realising, with their usual ignorance, what
they are doing). £ He drank like a fish . 3 ( £ He drank as a fish 3 would
mean, of course, when he was a fish.) So that one is forced, of course,
to say, £ He danced like a child 3 since £ as a child 3 would mean £ when
he was . 3 Meredith says ‘threading it with color, like yewberries the
yew . 3 33
Another linguist says: “In some other connexions like as a conjunc¬
tion may be 2 . 33 This linguist marked the above expression 4; he
marked item two 2 or 4, and item one he marked 2.
Still another, who marked this 4, says: “I dare say we shall have
to accept this too before long . 33
The various groups of judges agreed rather closely on this expres¬
sion. Their vote gives little support to those who consider this use of
like permissible.
Sentence Structure
I. Comma Splice
This book is valueless, that one has more to recommend it . 6
( 35 : established)
There was no other item on Ballot I that occasioned more un¬
certainty and disagreement than did this one. It is placed in this
section of established usages because the linguistic experts rated it
6 See also conclusions in the punctuation study in Current English Usage, p. 21 ft
DISCUSSION OF SPECIFIC ITEMS
115
high—twenty-two approving as against five who disapproved. No
other group of judges was so lenient; only a small majority approved
it. It is worth noting that the English teachers placed this lower than
did any other group, ranking it as 68 out of the hundred items. This
is perhaps due to their having found by experience that the comma
blunder is almost impossible to eliminate if even defensible excep¬
tions, like this, are allowed.
A further reason for disagreement among judges for this item proba¬
bly lies in the fact that the sentence, as it stands, with two clauses
in series, represents a different sort of problem from such an ex¬
pression as “The room was too cold, consequently we had to ad¬
journ.” If the latter had appeared on the ballot, it is possible that
many more judges would have agreed in condemning it. 7
II. Redundant He
My Uncle John, he told me a story. (199: illiterate)
A linguist points out: “For longer sentences this construction is
common in the best literary use from King Alfred on.”
In the sentence here given, this redundance is not in good use.
III. When in Definitions
Intoxication is when the brain is affected by certain stimu¬
lants. (129: disputable)
It is probably their weariness of hearing this day after day in the
classroom that caused the English teachers to vote nearly four to
one for inclusion of this expression among uncultivated usages. The
linguists were much more lenient, a majority regarding the locution
as admissible in colloquial English.
IV. I Read Where
I read in the paper where a plane was lost. (161: disputable)
A number of judges considered this expression to be good colloquial
English, but the majority would place it among uncultured usages.
< __
7 Mi. George Summey, Jr., in his study of punctuation {Modern Punctuation) ,
pp. 79-81, says: “In general, the comma is sufficient only when supported by
series, correlation, parallel form, climax, a common modifier, or the momentum of
the paragraph. . . . With no link work between successive statements the comma
is too light unless supported by special circumstances of structure or momentum.”
116 CURRENT ENGLISH USAGE
V. Incoherent Phrasing
1. Factories were mostly closed on election day. (68: estab¬
lished)
The linguists were nearly unanimous in approving this expression;
half the English teachers condemned it.
A majority of all the judges, then, consider this good colloquial
English. See also most anybody, in the section on pronouns.
2. Say, do you know who that is? (151: disputable)
Linguists say that this expression is not heard in England. In the
United States it is a dubious usage. Many people seem to feel that,
in special circumstances, say as a form of address is impertinence,
and condemn it.
Barbarisms and Improprieties
1. I wish he hadn’t of come. (221: illiterate)
This expression is indubitably illiterate.
m
2. If John had of come, I needn’t have. (224: illiterate)
This expression has no standing in current usage. See also hadn’t
of, above.
3. Hadn’t you ought to ask your mother? (222: illiterate)
All of the linguists, and most of the second jury, regard this locu¬
tion as illiterate.
4. I’ve no doubt but what he will come. (94: established)
A large majority of the judges approved this expression as ac¬
ceptable colloquial English, in spite of the purists’ violence of censure.
5. Reverend Jones will preach. (196: illiterate)
While the linguists condemned this locution by a majoi^ty of
four to one, the majority among the English teachers was only two
to one. Taking either judgment as a standard, however, this is not
an acceptable form.
DISCUSSION OF SPECIFIC ITEMS 117
6. They went way around by the orchard road. (103: estab¬
lished)
A British linguist says that this expression is “impossible in Eng¬
land.”
Thirty-eight of the forty-eight judges approve the expression as
acceptable in informal speech.
7. That there rooster is a fighter. (211: illiterate)
The judges were practically unanimous in rating this expression
as illiterate.
Idioms and Colloquialisms
1. It behooves them to take action at once. (6: established)
Nearly twice as many judges of all categories placed this expres¬
sion in 1 as placed it in 2 and 4 together. One linguist calls it
“antiquated,” another called it “hackneyed.” An author calls atten¬
tion to it§ use by Shakespeare, Ben Jonson, and Emerson. There was
more unanimity of opinion among the seven groups of judges on this
than on almost any other item, the standard deviation among the
ranks assigned being only 1.46.
This appears to be one of the few expressions belonging distinctly
to the literary and formal, rather than the spoken and informal,
language.
2. I had rather go at once. (7: established)
Several judges pointed out that in speech had rather is so pro¬
nounced (e. g. Td rather) as to be indistinguishable from would
rather.
The linguists ranked this as second among the hundred expressions
on Ballot I, but when the other judges’ rankings are taken into ac¬
count, the mean rank sinks to 16, with a standard deviation among
ranks bf over 14. This shows considerable disagreement among the
judges as to the exact place this item deserves to occupy. However,
each group of judges, except the authors, rated this high enough to
give it a place among the established usages.
118
CURRENT ENGLISH USAGE
3. You had better stop that foolishness. (10: established)
Every group of judges except the business men accepted this as
fully established. Linguists, editors, and English teachers rated it
highest.
4. My position in the company was satisfactory from every
point of view . (14: established)
The judges were nearly unanimous in rating this as entirely correct.
5. My viewpoint on this is that we ought to make concessions.
(67: established)
A British linguist remarks that this expression is “not used except
jocosely.”
There is considerable uncertainty manifested in the judges’ place-
ment of this; but the great preponderance of expert opinion ap¬
proves it as a good colloquial usage.
6. He toils to the end that he may amass wealth. (15: estab¬
lished)
«
There was great unanimity of opinion about this expression, no
group of judges ranking it lower than 5 out of the 100 expressions on
the ballot. Over half the judges classified it as belonging to formal
literary English. One or two remarked that it was rarely heard col¬
loquially.
7. In the case of students who elect an extra subject, an addi¬
tional fee is charged. (16: established)
Linguists says: “Trite but sound”; “I dislike this, but because it
is stylistically bad rather than that it is grammatically incorrect.”
8. I for one hope he will be there. (18: established)
This expression was almost unanimously approved as cultivated
colloquial English.
9. You may ask whomsoever you please. (22: established)
A large majority of the judges approved this as formal literary
English. A few condemned it on the ground that it was too stilted.
DISCUSSION OF SPECIFIC ITEMS
119
10. The women were all dressed up. (33: established)
Nearly ail the judges rate this as an acceptable colloquialism.
11. Take two cups of flour. (36: established)
The judges considered that the constant use of this expression in
standard works on cookery had established it as cultivated colloquial
and informal usage.
12. We have made some progress along these lines. (44: es¬
tablished)
Only one judge condemned this as illiterate; the expression is,
however, not fully approved, as witness these comments, by linguists:
“I dislike this, but rather because it is stylistically bad than be¬
cause it is grammatically incorrect.”
“A cant-phrase; slovenly argot of the loose-minded and semi¬
literate.”
“Avoided by the cultivated, but not illiterate.”
13. In hdpes of seeing you, I asked . . . (61: established)
The linguists gave a large majority for approval of this expression.
The English teachers would place it among disputed usages, demand¬
ing hope instead of hopes.
14. I didn’t speak to my uncle by long distance; I couldn’t get
through. (84: established)
This is a peculiarly British usage. About 25 per cent of the judges
classified the expression as “technical.” Most of the rest approved
it as acceptable colloquially. Cf. the American usage, below.
15. Haven’t you got through yet? (97: established)
(See also “I couldn’t get through,” above.)
Both the Standard and the New English Dictionaries define the
phrase, without comment, “get through with, to complete.” The New
Int. Diet, gives it as colloquial.
The expression puzzled some of the British judges, one of whom
says: “If on the telephone, ordinary; if equivalent to finished , not
120 CURRENT ENGLISH USAGE
English.” In England the expression could only mean, “Is your tele¬
phone connection completed?”
There is a clear majority for this as good colloquial American.
16. It is now plain and evident why he left. (92: established)
In each group of judges, there are some who approve this tautologi¬
cal expression as formal literary English, and others who condemn it
as illiterate. There are still, however, a majority who classify it as
cultivated colloquial usage.
17. My folks sent me a check. (100: established)
All the other groups of judges ranked this higher than did the
linguists.
Standard Diet.: “Widely used colloquially in spite of the drawing¬
room fastidiousness of some writers.”
New Int. Diet.: “Folk is now somewhat archaic; folks is commonly
considered colloquial.”
N. E. D.: “folks 4 pi.—The people of one’s family, parents, chil¬
dren, relatives.”
An American editor writes: “I have rated this 1 (formally cor¬
rect), thus displaying a good deal of hardihood. Whatever may be the
usage in England, in America the expression ‘his people’ does not
ring true. It is either upstage, or contains a covert sneer. No one ever
uses the phrase ‘my people.’ Perhaps the best way in ‘literary Eng¬
lish’ is to sidestep such a phrase altogether.”
This is evidently acceptable as colloquial.
18. I can't seem to get this problem right. (121: disputable)
A British linguist says: “Not English. Englishmen would use
don't."
A decided majority of the judges approved this expression as good
colloquial usage, but there were.enough who condemned it to make
its status doubtful.
19. Do you wish for some ice cream? (128: disputable) %
Remarks by linguists:
“Waitresses’ English.”
“Modem but now widely used by good authors.”
DISCUSSION OF SPECIFIC ITEMS
121
“Not classable, unidiomatic except in special circumstances.”
The vote of the English teachers would have placed this among
the established usages; the linguists are about evenly divided between
approval and condemnation.
20. Trollope’s novels have already begun to date. (132: dispu¬
table)
Dictionaries do not give this verb in this sense. Linguists disagree
widely: witness these comments:
“I do not know what this means.”
“Literary jargon.”
“Critics’ slang, perfectly good, even poetic.”
“Date, for example, is slang, in the strict sense of that term, and
a piece of slang that I personally dislike; but while I should certainly
not class it as ‘cultivated colloquial English,’ I hesitate to call it
‘vulgar English’ and so put it in the same category with ‘He won’t
leave me come in.’ ”
“Common slang among supposedly cultivated critics now!”
Perhaps this expression should be listed among those justified as
technical* The judges approved it, condemned it, and listed it as
technical, in nearly equal numbers.
21. It was good and cold when I came in. (142: disputable)
Of this expression, linguists say:
“Uncultivated; if nice were substituted for good I would assign
this to 2 (cultivated colloquial usage).”
“This is used widely in my section; I marked it 2 ; probably it
could be marked 1.”
“This seems to me idiomatic —good probably in the sense of ap¬
proaching perfection, thoroughly ”
Speech teachers place this among established usages; the other
judges classify it as disputable. We are certainly not justified in
considering this expression as definitely bad and to be stamped out.
22. The British look at this differently than we do. (168: dis¬
putable)
A British linguist remarks, somewhat heatedly: “Good as literary
or formal (but wrong for colloquial use). ‘From, to, than, all in best
122 CURRENT ENGLISH USAGE
authors’—Concise Oxf. Diet. Differs plus dative (Tacitus and else¬
where. Tell the purists Tacitus was a Roman historian and Latin
was Ms native language). Different: unlike . Like to Is a literary usage;
why not unlike to? The reference to differ is as superficial as most
puristic rubbish (speaking dispassionately). The logical analogies of
opposite to } contrary to, dissimilar to , would never occur to these bone-
heads. However, there are a number of cases where different to can
mean something different. E. g. £ He is quite different to me’ (He
behaves towards me quite differently). So that there is a fairly sound
non-purist reason for different from, and I try to say it. Different
than is good formal English, but in colloquial English I think one
would say differently to us (or from us), rather than than we do”
Another linguist says: “Continued observation has convinced me
that different than is in the best of colloquial and literary use, in
spite of the purists. In many cases from Is decidedly awkward.”
This expression has an astonishing range in judges’ placements.
The business men ranked it 20 out of 102; the linguists, 73, with
the other groups distributed in between. Over 15 per cent of the
judges approved it as formal English, while about 50 per cent con¬
demned it as illiterate. In short, where experts disagree so widely, it
will be unsafe for others to be dogmatic about the standing of this
expression.
Changes in Definition and Use of Words
r. Why pursue a vain hope? (3: established)
This expression was almost unanimously approved, a large majority
of judges placing it among expressions appropriate to formal or liter¬
ary English.
2. The defendant’s case was hurt by this admission. (17: es¬
tablished)
A large majority of the judges approved this as cultivated collo¬
quial English; not one condemned it.
1
3. The honest person is to be applauded. (23: established)
Votes of the judges were nearly equally divided as approving this
expression as formal or as colloquial English.
DISCUSSION OF SPECIFIC ITEMS
123
4. We got home at three o’clock. (27: established)
This expression was placed in class 1 by thirty-eight judges; 159
placed it in class 2.
This expression is clearly good English for colloquial, rather than
literary, use. The business of English teachers so far as usage is con¬
cerned is evidently to strive to establish cultivated colloquial usage;
so there is no reason why any teacher should waste time in trying to
suppress this or any other good colloquial expression.
5. I’ve absolutely got to go. (54: established)
The linguists rated this highest, the business men lowest. Eleven
out of fifty English teachers condemned the expression—perhaps a
survival of the admonitions they received when they were in grade
or normal school training. The N. E. D. lists it as “recent colloquial.”
Some British judges considered it an American locution.
One linguist says, “Acceptable if it means ‘am compelled by au¬
thority or circumstances. 7 77 Another says, “Acceptable if it means
‘can’t resist inclination. 7 77
This is an acceptable colloquialism.
6. I have got my own opinion on that. (87: established)
Standard Diet.: “Colloquial. 77
New Int. Diet.: “Pleonastic. 77
N. E. D.: “In familiar language. 77
A linguist remarks: “Quite different in meaning from ‘I have my
own opinion. 7 ‘He has black eyes, 7 but ‘He’s got a black eye. 7 77
Linguists rank this highest, business men, lowest. There is a clear
majority for approval as colloquial.
7. Leave me alone, or else get out. (140: disputable)
The authors, in most cases the most meticulous among the judges,
rated this expression higher than did most of the other groups.
According to the New English Dictionary this expression is
synonymous with “let me alone 77 and is quite accepted. This is
confirmed by the judgment of the British linguists in the present
study; their composite rank for it is 2.1. Contrast with next item.
124 CURRENT ENGLISH USAGE
8. He won’t leave me come in. (217; illiterate)
British linguists reject this expression as unfamiliar to them; one
says that it is heard in Irish dialects. About 98 per cent of the judges
disapproved it. Contrast leave me alone, above.
9. In this connection, I should add . . . (8: established)
Comments:
“I dislike this, but rather because it is stylistically bad than be¬
cause it is grammatically incorrect. 57
“Trite but sound. 57
10. He has no fear; nothing can confuse him. (28: established)
Some judges suggested that perhaps the word “confuse 57 was ill-
chosen; the majority approved the expression as correct.
ix. I drove the car around the block. (38: established)
This was almost unanimously approved as cultivated colloquial
English.
12. The New York climate is healthiest in fall. (40: estab¬
lished)
No linguist disapproved the expression; about one-third of the
English teachers, probably influenced by the condemnation visited
on this construction by most handbooks and rhetorics, condemned it.
The preponderance of opinion is clearly in favor of its approval as
cultivated colloquial English, the handbooks to the contrary not¬
withstanding.
13. One is not £t to vote at the age of eighteen. (42: estab¬
lished)
Here the two groups of judges agreed that this expression is per¬
fectly acceptable colloquial English.
*
14. I can hardly stand him. (51: established)
A linguist says: “Here the rest of the context must be considered.
‘Stand him 5 strikes me as general, but not cultivated colloquial;
DISCUSSION OF SPECIFIC ITEMS
125
while Tew can stand prosperity 5 is undoubtedly good colloquial,”
A very large majority of the judges approve this as cultivated col¬
loquial usage.
15. There are some nice people here. (57: established)
N. E. D.: “In common use from the latter part of the 18 th century
as a general epithet of approval or commendation. 55
A business man says: “I think the word nice is decidedly overdone.
On the other hand, I classify this as 2 because it seems to me justified
by usage. 55
A linguist says: “I am so accustomed to hearing well bred people
use nice that I can hardly escape using it in formal writing; I expect
to see the word rated l. 55
Another: “I confess a personal liking for nice. I do not see why
exactly or precisely may not be used in the sense of exquisite when
applied to human beings or highly organized objects. I believe that
the word nice has a future. 55
The judges of all groups displayed practical unanimity in approv¬
ing this use of the word “nice 55 as cultivated colloquial English.
%
16. His attack on my motives made me peevish . (64: estab¬
lished)
It is interesting to note that nearly one-third of the judges ap¬
proved this expression as appropriate to formal literary English, while
one-seventh of them condemned it as illiterate. About one-half clas¬
sified it as colloquial.
This is clearly established as acceptable English.
17. He worked with much snap. (71: established)
Only two judges approved this as formally correct English; most of
the remainder, however, considered it as good colloquial usage.
18. This room is awfully cold. (72: established)
One linguist says: “I say this, but I think I never write it. 55 His
opinion is evidently concurred in by most of the judges, for only 4
approve the expression as formally literary, whereas 183 approve
it as colloquial. 30 condemn it as illiterate. Dictionaries disagree, some
126
CURRENT ENGLISH USAGE
regarding it as slang, others as colloquial. It probably belongs among
those expressions which are emerging from slang into the lighter
levels of cultivated speech, and is certainly not worth trying to elimi¬
nate from the speech of school children.
19. I have a heap of work to do, (78: established)
About one-third of the English teachers classified this expression
as illiterate; the other two-thirds, together with all but one of the
linguists, approved it as good colloquial usage.
20. He made a date for next week. (88: established)
All but two of the linguists approved this expression as acceptable
colloquially; nearly half the English teachers condemned it. The ex¬
pression seems to have emerged from the level of slang to that of
accepted informal speech.
21. Can I be excused from this class? (96: established)
In order to keep the scheme of classification consistent? this is
placed among the established usages on the basis of the rating of the
linguists, more than three-quarters of whom approved the expression
as colloquial. Its position here is, however, made somewhat dubious
by the much lower ratings bestowed by the other groups of judges.
Comments by linguists:
“If the speaker means may , this is illiterate. The question might be
one of possibility.”
“Can is often condemned, but is common in our best writers.”
Probably the fitness of this expression is a matter of taste, rather
than usage. But it cannot be listed as vulgar or uncultivated in the
face of the large number of judges who recognize its frequent use
by cultivated people.
22. Is your insurance sufficient coverage for your house?
(107: established as technical) <-
This was the only expression on either ballot which was unquali¬
fiedly classified as technical by the judges. As such, and in its own
field, it is entirely correct usage.
DISCUSSION OF SPECIFIC ITEMS
127
23. That clock must be fixed. (108: disputable)
British linguists remark that this expression is distinctively Ameri¬
can; several point out that in England this could only mean “fixed
to the wall, or fastened into position.” The Standard, New Interna¬
tional, and New English Dictionaries all characterize as colloquial
the use of the word represented in this sentence. Linguists and mem¬
bers of the M. L. A. ranked this lowest; speech teachers and editors
placed it among entirely acceptable expressions.
This expression is evidently not at present safely established, but
it cannot be called a solecism.
24. Have you fixed the fire for the night? (59: established)
British linguists remark that the verb fix, in England, is used only
in the sense of make fast.
Practically all the judges approve this expression as cultivated
colloquial usage, certainly for the United States.
25. The Rock Island depot burned down last night. (114: dis¬
putable)
The value of the testimony on this expression is probably weakened
by the fact that the sentence was not worded so as to make it un¬
mistakable that depot referred to a railway station.
All dictionaries give this meaning of the word as being peculiar
to the U. S. The judges for the most part agreed, and a large majority
(though fewer than 75 %) approved it as correct usage.
Comments:
“Incorrect, if depot means railway station.” (British linguist)
“Anglicism? My grandmother, English and an actress, always said
‘depot. 7 77 (Speech teacher)
“Depot seems to have gone out of fashion very rapidly. Twenty
years ago everyone used it in my home town without question. 77
(Linguist)
“Station is gaining. 77 (Speech teacher)
“This was once good American usage. It is now countrified rather
than wrong. 77 (Editor)
Depot is probably still quite correct in the United States, but seems
to be going out of fashion.
128 CURRENT ENGLISH USAGE
26. He went right home and told his father. (118: disputable)
The dictionaries record no objection to this use of right; the
N. E. D. gives many citations from English literature to support the
usage.
Each group of judges except the linguists and the authors placed
this among established usages.
One British linguist says: “Incorrect, if right means straight.”
But another remarks: “Good colloquial usage if it means ‘the whole
w r ay home. 7 Although straight home means immediately (statim)
home , straight can’t be used in this sense in any other connexion,
as far as I know. Straight home really means ‘starting at once and
taking the shortest route. 7 77
This is probably entirely correct for informal speech.
27. I expect he knows his subject. (120: disputable)
A linguist says: “I hesitate about this. I use it, but against my own
convictions. 77
The Standard and New International dictionaries call this ex¬
pression “a colloquial solecism. 77 The N. E. D. says: “Now rare
in literary use ... Its misuse is often listed as an Americanism, but
is very common in dial., vulg., or carelessly colloquial speech in Eng¬
land. 77
One-third of the linguists, and about half the other judges, con¬
demn this expression as uncultivated; its standing is at best un¬
certain.
28. I guess Ill go to lunch. (117: disputable)
Standard Dictionary: “A colloquialism, esp. in the northern U. S.,
but occurring in English literature as early as the 17 th century. 77
N. E. D.: “Colloq. in northern U. S. 77
Rupert Hughes says: “J guess as used in America is classically
good English but is offensive to the British who have let it grow
obsolete. Galsworthy, Wells, and others simply cannot reproduce our
usage of it in their ridiculous efforts at American dialect.
“As I said in a paper quoted in Mencken 7 s American Language,
we are under no obligation to accept orders from our English cousins
about our common heritage, but I guess is perilous to use because it
arouses controversy instead of understanding. 77
DISCUSSION OF SPECIFIC ITEMS
129
British linguists say: “Illiterate here; in U. S. acceptable as col¬
loquial.” “This strikes me as American. Used facetiously, as such, by
English people.” “As proper as 1 fancy, 1 imagine, 1 think.”
Of twenty-three authors, twelve approved this expression as liter¬
ary English, eleven condemned it as illiterate. A large majority of
the remaining judges classified it as acceptable in informal American
English. It is probably still safely established in the United States.
29. I calculate to go soon. (201: illiterate)
This is a localism, and is not allowable in general use.
30. That boy’s mischievous behavior aggravates me. (162: dis¬
putable)
American dictionaries label this expression “colloquial”; the Ox¬
ford Dictionary calls it “familiar.” The judges are nearly equally
divided between approval and condemnation. While not altogether
incorrect, it is evidently not established as a cultivated colloquialism.
31. It Is liable to snow tonight. (170: disputable)
The Oxford Dictionary supports the expression with many cita¬
tions from writings from 1682 to 1896 .
The fact that nearly as many judges approved this expression as
condemned it places it among usages concerning whose correctness
nothing positive can be said on either side.
32. They went in search for the missing child. (171: disputa¬
ble)
Both groups of judges voted about two to one against the in¬
clusion of this expression among permissible usages.
33. John was raised by his aunt. (173: disputable)
Linguists and dictionaries agree that this expression is U. S. dialect,
especially in the South and West. English teachers and business men
rate it relatively high. W T hile not commonly accepted as educated
usage, it cannot be said to be a solecism.
130 CURRENT ENGLISH USAGE
34. Both leaves of the drawbridge raise at once. (208: illiter¬
ate)
Although the English teachers rate this expression more leniently
than do the linguists, both groups give large majorities for its in¬
clusion among illiterate usages.
35. The party who wrote that was a scholar. (198: illiterate)
A linguist says: “Not classable—strictly correct but unidiomatic.”
Nine judges rated this as a technical expression, probably because
they felt that it was allowable in legal jargon. As non-technical
English, it is not in good standing.
36. I was pretty mad about it. (122: disputable)
A linguist remarks, “Hardly English. English colloquial sick”
Standard Dictionary: “. . . archaic in lit. Colloquially, in the
U. S., mad in this sense is very common, and as a provincialism is
not uncommon in England. Its use may be regarded as permissible
colloquially when connected with a cause of vexation that is not a
person. 5 ’
N. E. D.: “Now only colloq.”
In the U. S. this is probably acceptable in informal discourse.
37. That will be all right, you may be sure. (48: established)
There is practically no disagreement among the judges in classify¬
ing this expression as established in cultivated colloquial usage. Only
five out of over 200 condemn it as vulgar. Some of those who classed
it as colloquial indicated that they believed it would soon belong with
literary English.
The N. E. D. gives 1686 as the date of its earliest recorded use, and
makes no objections to the form.
Debated Spellings
Such naif actions seem to me absurd. (148: disputable)
The Standard Dictionary says: “Same as naive”
One linguist says: “In English usage naive is employed without
DISCUSSION OF SPECIFIC ITEMS 131
reference to gender. The use of naif for naive is a purist affectation.”
Another remarks: £ T have read such sentences a few times.”
18 per cent of all the judges classified this expression as technical;
the rest were almost evenly divided between the other three classifica¬
tions—almost as many thought it acceptable for formal literary Eng¬
lish as condemned it as illiterate! The expression then, clearly belongs
among disputable usages because there is no agreement as to its
standing.
_0?3 ill <?£l
PRACTICAL CONCLUSIONS AS TO GRAMMAR
AND USAGE
What practical conclusions applicable In classroom instruc¬
tion can be drawn from the findings of this grammatical study?
As In the case of punctuation, two lines of procedure suggest
themselves: usages upon which the judges strongly agree can
be profitably taught; in regard to usages upon which the judges
are evenly divided, dogmatism is unjustified. Extensive drill on
either form of a divided usage would clearly be a loss of time;
and it is equally obvious that no class time should henceforth be
wasted In an effort to eradicate any construction here listed as
established—no matter what the personal preference of the in¬
structor or the dictum of the adopted text.
Two considerations guide the teacher of composition In his
approach to the subject of grammatical usage: first, how near
to correctness and clarity he can bring the average language level
of his class; and second, what he can do to lift superior students
from mediocrity to elegance in their use of English. Where com¬
position classes are sectioned according to language ability, the
problem is somewhat simplified, since in superior sections the
more illiterate mistakes in sentence structure and usage will not
need to be eradicated and stress can be placed on finer points of
style. In the mixed class, however, one has constantly to*choose
between the presentation in group instruction of a standard far
more meticulous than can be attained by the average student, or
the presentation of a standard attainable by the average but
132
CONCLUSIONS
133
little if at all higher than that already reached by the better stu¬
dents. The wise Instructor will probably solve the difficulty by
drilling the class upon the requirements to be demanded of the
average and either using the superior students as assistants In
carrying on this drill, or excusing them entirely from participa¬
tion therein, and then presenting to them individually or in a
separate group the more exacting standards to which he thinks
they may attain . 8 Whichever of these methods he adopts, he will
certainly in marking themes accept from the average student
any usage classed in this study as established or disputable. To
the student more skilled in the use of English, he should privately
suggest more elegant locutions in place of disputable usages. The
superior child capable of reaching distinction in speech or writ¬
ing deserves such individual instruction as will open up to him
the finer manipulation of language. But except in those limited
portions of the composition course necessarily devoted to class
drill or individual conference on items of grammar and usage,
there is little doubt that both average and superior students alike
will profit more by attention to the interest and clarity of their
oral and written work, to the richness of their observation of life,
the soundness of their thought, the organization of their ma¬
terial, and the originality of their expression, than by the most
thorough and painstaking usage drill.
Upon the moot question of how far the study of formal gram¬
mar can improve the speech habits of our students, the present
study throws little light. To the study of grammar itself, how¬
ever, it makes anew two* contributions of major and perennial
importance.
In the first place, grammar is seen to be not something final
and static but merely the organized description or codification
of tfie actual speech habits of educated men. If these habits
change, grammar itself changes, and textbooks must follow suit.
8 Class Size in High School English. Dora V. Smith, University of Minnesota
Press.
134
CURRENT ENGLISH USAGE
To preserve In our textbooks requirements no longer followed
by the best current speakers Is not grammatical but ungram¬
matical. It makes of grammar not a science but a dogma.
Many teachers —and, for that matter, most persons who have
not specialized in the observation of language ways—conceive
of rules In grammar as laws to which language must conform, In
the same way that the man In the street conceives of physical
laws as governing the behavior of matter. We are accustomed to
saying, for instance, that certain laws “govern” the behavior of
gases under pressure, and we naturally think of the operation
of these laws as being like that of the laws which govern the
behavior of a civilized individual in a commonwealth—as some¬
thing promulgated to regulate action. The physical law is really
a statement of how gases have been observed to behave under
certain conditions, and the physicist stands ready to change the
“law” the moment observation shows this behavior to be in any
way different from what he had formerly thought. In the same
way, the grammatical rule that the complement of the verb to be
is always in the nominative case, is also merely a statement of
the way people actually write and speak, and the moment people
cease to write or speak in this way, this particular “law” of gram¬
mar must be changed. In the paragraphs numbered 2 and 3 ,
immediately below, are cited certain apparent violations of the
rules of formal grammar which have become so well established
in the language of educated people that they are now in reputable
usage. The use of the word violation in this connection is really
highly inaccurate, in the same way that It would be inaccurate
to say that a gas’s failure to act in the way predicted by Boyles’
Law was a violation of that law. We should say, instead, that
the statement of the law must be revised so that it would more
accurately describe the behavior of the gas. *
There are three such tentative general revisions of the gram¬
mar of written and spoken English which this study seems to
validate.
CONCLUSIONS
135
1 . A number of usages entirely in accord with the present rules of
formal grammar are apparently avoided by careful speakers and
writers because they are regarded as finical or pedantic. Among
these 9 are the use of the article an with certain words (such as
historical) beginning with h; the strained avoidance of the split
infinitive; and insistence upon a formal sequence of ones in such
a sentence as “One must mind one’s manners.” These expressions
we should not forbid; but we certainly should not encourage their
use by dogmatic requirement.
2 . There are expressions which are condemned by most handbooks
and which are listed among improper usages in the chapters on
diction in many school rhetorics but which are nevertheless in fre¬
quent use by educated speakers. It might be wise not to assign
such chapters to pupils until the acceptability of the expressions
has been checked by the findings of this study. 10
3 . Formal grammar is apparently at fault in setting up rigid rules
for the case of personal pronouns after to be and of the interroga¬
tive pronoun who . 11
The second contribution made by this investigation of the
study *of grammar relates to the principle which apparently
guides grammatical change. If the point of view of the judges in
reaching their decisions regarding these usages is correct, the
governing principle in the use of language, as of punctuation, is
clarity of thought. “Which phrasing most accurately expresses
the intended meaning?” seems to be their constant question.
The usage which does this will be the usage upon which educated
persons finally settle and which will thus become grammatical.
If meaning is the midwife at the delivery of usage (of which
grammar is only the codified description), should not meaning
likewise be the governing principle in the teaching of formal
grammar? Probably no study is allotted more time and is more
barren of results than that from which our grammar schools de-
9 See index under hyper-urbanisms.
10 See index for not — as; reason why; none are; healthy for healthful; pretty
good; back of; the use of shall, will, should, would, etc.; try and; got to; the
split infinitive; slow and other adjective forms used as adverbs (see index under
adverbs ); fix for repair; the position of only, etc.
11 See index under pronouns, case of.
136 CURRENT ENGLISH USAGE
rive their name. After five or six years of grammar work in the
elementary schools; after endless diagramming (which is but
parsing in pictorial reincarnation); after painful memorizing of
rules and definitions; and after constant composition of illustra¬
tive exercises, many a high school freshman cannot write or
speak a decent English sentence. Propose a review of grammar,
and the response will be a politely smothered groan. Yet often
these same children will enjoy the grammar work incidental to
foreign language, and will suddenly awaken to facts about their
own tongue which all their study of English grammar never re¬
vealed. The reason for this is patent. Although the teaching of
other grammars has its vices, it has usually the virtue of being
purposeful. Grammar is felt to be an aid to expression in the
new language—a way of getting thoughts marshaled and in¬
telligibly put. It is something dynamic, a usable tool.
All too often the teacher of English grammar is not construc¬
tive but analytic. Instead of doing something interesting by
means of grammar, the pupils are asked to pull the language to
pieces and examine its dead fragments. But children are no
analysts. They are not interested in pieces; they want moving
wholes. The result is that instead of seeing that grammar is
merely common sense applied to language to help us express our
ideas clearly, children regard it as a jumble of abstract rules,
quite meaningless and negligible in every day life. Most of us
can recall learning as children from a text still widely employed
that a sentence was a thought expressed in words. That was the
first and last time thought was mentioned in the book. The rest
of the discussion was about words. We named and classified
words; we declined and conjugated them; we juggled with their
gender; we stuck them together in what we called sentences;
we hacked real sentences up into words; we isolated poor shiver¬
ing little if s and tke’s and examined them mercilessly in all their
nakedness. And we naturally concluded that sentences were
mere mosaics plastered together with grammatical cement. That
CONCLUSIONS
137
the sentence was a living organism, an indivisible whole, the
throbbing body of an interesting thought, never occurred to us.
The whole method of attack was pedagogically wrong. Little
brains and little fingers are very much alike; both lack the
power of fine manipulation. The baby can grasp a squirming
kitten; he fumbles over and drops a tiny bead. But textbooks on
grammar are written by grown ups who have at once the power
of handling detail and the understanding of wholes, neither of
which the child has yet acquired. They present a subject not
descriptively as youth demands but analytically as age prefers.
They contain the anatomy and not the biology of language. They
arrest language instead of putting it in motion.
If it be that some study of grammatical laws is necessary to
mature manipulation of language, the study should begin at the
other end. The whole sentence should first command attention.
Children will be quick to recognize its completeness or incom¬
pleteness when they are looking at it as a statement of an inter¬
esting thought, a deed performed, a fact noted. If the sentence
must be cut up at all, let it be into big thought blocks. If dia¬
gramming must be done, let it be the old fashioned sausage-link
kind, expressing simply, without hair splitting, the big primary
grammatical relations.
There is undoubtedly a place in the curriculum for a thorough
study of those grammatical principles which seem to govern all
language because they also govern the logic of thought, and
hence of its communication. But unless this study is a study of
logic and not of formal rules; unless this study does keep pace
with actual usage instead of insisting upon a petrification of
principles which reduces the grammar text to a volume of folk
lore and curious myths, grammar study can neither change il¬
literate usage nor produce that mature power over the manipula¬
tion of language which a knowledge of fundamental principles
gives the scientist or the artist over the manipulation of the
materials of his science or his art.
INDEX
A orange, 56, 98
Abbreviations, 26
Adjectives
as adverbs
bad, 44, 101; fine, 45, 101;
good, 53, 102; likely, 44, 101;
noble, 57, 102; pretty, 28, 99;
quick, 35, 100; real, 46, 101;
slow, 29, 36, 99, 100; sure,
41, 48, 100, 101
comparison
worse, 43, 107
pronominal
these, 48, 97; this, 31, 97;
those, 24, 97
superlative for comparative, 42,
106
Adverbs
badly, 36, 103; very with past
participle, 43, 102; way, 31,
117
comparison
all the further, 54, 106
form.
See adjectives used as adverbs
position of
not only, 44, 105; only, 35,
104
Aggravate, meaning, 45, 129
Agreement
pronoun and antecedent
everyone . . . they, 38, 74
subject and verb
kind . . . are, 30, 89; one of
^(plural noun) were, 56, 89;
there was, 37, 89
Ain’t I, 48, 95
Ain’t, with third-person subject, 54,
96
All right, 24, 130
All the further, 54, 106
American Colloquial English, 17, 19
American Literary English, 17, 19
Among with collective, 56, 110
An preceding h 7 22, 98
Applauded, 23, 122
Archaisms, 17, 19
Aren’t I, 47, 90
Around, preposition, 38, 108
Article
indefinite, 22, 98
a orange, 56, 98
omitted
Reverend, 49, 116
As
as well or better than, 39, 113;
as ... as, 27, 111; as re¬
gards . . . , 34, 108
Awfully, meaning, 30, 125
'Awoken, 53, 87
Back of, 24, 109
Bad, as adverb, 44, 101
Badly after felt, 36, 103
Begun, past tense, 54, 84
Behooves, 22, 117
Between
each, 44, 107
of more than two, 36,107
Busted, 47, 96
But
after can’t help, 44, 98
but what, 37,116
Calculate, meaning, 54, 129
Can, as permissive, 37, 126
Can’t help but, 44, 98
139
140
INDEX
Can’t seem, 40,120
Case
in the case of, 23, 118
Case. See Nouns, Pronouns
Classification, basis of, 15 ff.
Colloquial English, 3, 17, 19, 20
Colloquialisms, technical, 25; see
also Idioms and colloquialisms
Come, past tense, 56, 86
Comma splice, 34, 114
Comparison
further, 28, 105
Complected, 56, 98
Conclusions, 61, 132-137
Concord. See Agreement
Confuse, 23, 124
Conjunctions and conjunctive ad¬
verbs, 30
not as ... as, 27, 111; reason
why, 28, 111; so, 23, 111
Coverage, 24, 126
Cups of flour, 25, 119
Current English Usage
dictionaries, attitude toward, 13-
15
distribution of judges’ rankings
in, 11
technique, 2-4
Data, as singular, 49, 70
Date
meaning, 36, 126
verb, 41, 121
Depot, meaning, 39, 127
Dialect, 17, 19
Dictionaries, 13, 16
Differently than, 45, 121
Disputable usages, 33-51
Don’t with third-person subject, 45,
46
Double negative
can’t help but, 44, 98; hardly,
53, 99; haven’t but, 43, 98
Dove, past tense, 41, 86
Drank, past participle, 55, 87
Dress up, 28, 119
Drove the car, 25, 124
Drunk, past tense, 54, 85
Due to, 47, 93
Eat, past tense, 47, 86
Either with three, 40, 112
Established usages, 22-32
Evenings, adverbial genitive, 38,
110
Everybody . . . their, 44, 74
Everybody’s else, 44, 81
Everyone . . . they, 38, 74
Expect, meaning, 40, 128
Factual sources, used in word study,
15-17
Farther. See Further
Fine, adverb, 45, 101
First, position of, 39, 110
Fit, meaning, 28, 124
Fix, verb, meaning, 29, 38, 127
Folks, 38, 120
For one after I, 23, 118
From whence, 34, 110 *
Functional change
nouns to verbs
loan, 24, 30, 97; price, 30, 96;
taxi, 25, 96
Further vs. farther, 28, 105-106
Gerund, case with, 37, 40, 42, 94-95
Get through, 25, 39, 119
Good, adverb, 53, 102
Good and cold, 43, 121
Got
meaning, 36, 123
’ve got to go, 29, 123
Got home, 33, 123
Gotten, past participle, 39, 87
Guess, meaning, 39, 128
Had better, 27, 118
Hadn’t of, 57, 116
Hadn’t ought, 57, 116
Had of, 57, 116
INDEX
141
Had rather, 27, 117
Haven’t but, 43, 98
Hardly, in double negative, 53, 99
He
after infinitive to be, 41, 80
redundant, 54, 115
Heap, meaning, 36, 126
Healthiest, 28, 124
Him
after form of to be, 42, 44, 78
after than, 48, 79
His or her, 23, 73
Home, adverb, 28, 109
Hopes, after in, 35, 119
Hurt, 23, 122
Hyper-urbanisms, 22-23
I, in predicate after form of to be,
22, 77
Idioms and colloquialisms
had better, 27, 118; had rather,
27, 117; behooves, 22, 117;
can’t seem, 40, 120; cups of
flour, 25, 119; date, verb, 41,
121; differently than, 45,121;
dress up, 28, 119; folks, 38,
119; get through, 25, 39, 119;
good and cold, 43, 121; I for
one, 23, 118; in hopes of, 35,
119; in the case of, 23, 118;
plain and evident, 37, 120;
point of view, 23, 118; to the
end that, 23, 118; viewpoint,
35,118; whomsoever, 23,118;
wish for, 41, 120
If for whether, 30, 112
Illiterate as defining label, 52
Illiterate usages, 52-60
In back of, 39, 109
In front of, 23, 109
In search for, 46, 129
In this connection, 23,124
Infinitive
case of pronouns after. See Pro¬
nouns
split, 22, 29, 91-92
It, as impersonal pronoun, 35, 72
Judges
instructions to, 66-67
personnel, 3, 66
Kind . . . are, 30, 89
Kind of, 41,103
Laid, intransitive verb, 24, 82
Lay, intransitive verb, 55, 57, 82
Leave for let, 42, 56, 123-124
Leonard, S. A. See Current English
Usage
Levels of usage, 3
Liable, meaning, 45, 129
Like
for as, 46, 48, 114
for as if, 47, 113-114
Likely, adverb, 44, 101
Literary English, 3, 17-20
Loan, verb, 24, 97
Mad, meaning of, 40, 130
Me
after is, 35, 77
after than, 40, 79
Meaning
aggravate, 45, 129; all right, 24,
130; applauded, 23, 122; aw¬
fully, 30, 125; calculate, 54,
129; confuse, 23, 124; con¬
nection, 23, 124; coverage,
24, 126; date, 36, 126; depot,
39, 127; drove, verb, 25, 124;
expect, 40, 128; fit, 28, 124;
fix, verb, 29, 38, 127; got, 29,
123; got home, 33,123; guess,
39, 128; have got, 36, 123;
healthy, 28, 124; heap, 36,
126; hurt, 23, 122; leave, 42,
56, 123-124; liable, 45, 129;
mad, 40, 130; nice, 29, 125;
INDEX
142
Meaning (cont.)
party, 54, 130; peevish, 24,
125; pursue, 23, 122; raised,
46, 129-130; right, 39, 128;
snap, 30, 125; stand, verb, 28,
124
Most for almost
adjective, 45, 82
adverb, 46, 103
Mostly, 35, 116
Much as pronoun, 23, 82
Myself for me, 37, 71
Naif, spelling of, 43, 130
Neither, 41, 75
Neither . . . are, 47, 75
Nice, 29, 125
Noble, adverb, 57, 102
None, member of, 34, 73-74
Not only, position of, 44, 104,
105
Nouns
case
Pike’s Peak, 24, 70; ways, 47,
71; with gerund, 37, 40, 42, 94-
95
number
data, 49, 70; woods, 40, 70;
works, 34, 70
Off of, 46, 111
One
one . . . he, 38, 76
one . . . one . . . one, 22, 76
Only, position of, 35, 102
Participles
use of
providing, 36, 93
Party, meaning, 54, 130
Peevish, 24, 125
Place names
punctuation of
Pikes Peak, 24, 70
Plain and evident, 37,120
Point of view, 23, 118
Prepositional phrases
in back of, 39, 109; in front of,
23, 109
Prepositions
around, 38, 108; between, of
more than two, 36, 107; be¬
tween each, 44, 107; for after
want, 56, 92; to, 55, 108
omitted, 28, 109
back of, 24,109; evenings, 38,
110; home, 28, 109
redundant
at after where, 49, 111; off of,
46, 111
unusual uses of
under, 23, 108
used as conjunction
like for as, 46, 48, 114; for as
if, 47, 113-114
Pretty, adverb, 28, 99
Price, verb, 30, 96
Pronouns
much, 23, 82 r
case
everybody’s else, 44, 81; he
after infinitive to be, 41, 80;
him after form of to be, 42, 44,
78; him after than, 48, 79; I
after form of to be, 22, 77; me,
after is, 35, 77-78; me after
than, 40, 79; them after in¬
finitive to be, 43, 80; they after
infinitive to be, 39, 79; us after
form of to be, 38, 78; who, as
object, 35, 80-81; whoever, as
object, 36, 80; whom for who,
53, 80-81; she as object of a
preposition, 54, 81; with ger¬
und, 40, 42, 94-95
gender, 23, 73
which, 55, 73-74; whc&e, 27,
73
impersonal
it, 35, 72; one, 22, 38, 76-77;
they, 27, 31, 72; you, 30, 72
INDEX
143
intensive
myself, 37, 71; yourself, 43,
72
number
everybody . . . their, 44, 74-
75; everyone, 38, 74; neither,
41, 47, 75; none, 34, 73-74
reference
that, 23, 75; they, 76; which,
24, 76
relative omitted, 27, 72
Proven, past participle, 38, 88
Providing, 36, 93
Pursue a vain hope, 23, 122
Quick, adverb, 35, 100
Raise, intransitive verb, 55, 129-
130
Raised, meaning, 46, 129-130
Ranking by linguists, 4-11
Real, adverb, 46, 101
Reason . . . was because, 31,112
Reason why, 28, 111-112
Redundancy
(John) he, 54, 115; from
whence, 34, 110
Reference. See Pronouns
Reverend, without article, 49, 116
Right, as adverb, 39, 128
Run, past tense, 58, 85
Say, as interjection, 44, 116
Says, historical present, 55, 87
Sentence structure
comma splice, 34, 114
Set, intransitive verb, 58, 83
Shall vs. will, 23, 83
She, as objective of a preposition,
54, 81
Should vs. would, 29, 84
Slow, 29, 99-100
Snap, meaning, 30, 125
So after not, 23, 111
Some after going, 42, 104
Some for somewhat, 46, 104
Stand, verb, meaning, 28, 124-125
Sung, 48, 85
Superlative for comparative of ad¬
jectives, 42, 106
Sure, as adverb, 41, 48, 100-101
Swang, past tense, 58, 86
Taxied, verb, 25, 96
Technical usage, 24
Tense
sequence, 43, 88
says, 55, 87
See Verbs
Than him, 42, 79
Than me, 40, 79
That
indefinite antecedent, 23, 75
reference of, 75
that there, 55, 117
Them, after infinitive to be, 43, 80
There was with compound subject,
37, 89
These kind, 48, 97
They
after infinitive to be, 39, 79
as impersonal pronoun, 31, 72
reference of, 76
This
pronominal adjective, 31, 97
this much, 23, 82
Those, pronominal adjective, 24, 97
To no sense, 55, 108
To the end that, 23, 118
Try and ... , 34, 92
Under these circumstances, 23,
108
Us after form of to be, 38, 78
Verb formation
prejudices in respect to, 59-60
144
INDEX
Verbs
agreement with subject
aren’t I, 47, 90; it don’t, 45,
90; kind . . . are, 30, 89;
Martha don’t, 46, 90; singular
subject modified by phrase, 45,
89; one of (plural noun) were,
56, 89; there was with com¬
pound subject, 37, 89; you
was, 56, 91
conditional, 29, 84
mood
was, after wish, 37, 89;
wasn’t, 30, 88
participles
due to, 47, 93
tense, future
shall, 23, 24, 83-84; will, 41-
42, 83-84
tense, past
begun, 54, 84; come, 56, 86;
dove, 41, 86; drunk, 54, 85;
eat, 47, 86; run, 58, 85; sung,
48, 85; swang, 58, 86
tense, past participle
awoken, 53, 87; drank, 55,
87; gotten, 39, 87; proven,
38, 88
tense sequence, 27, 88
transitivity
laid, 24, 57, 82; lay, 55, 82;
raise, 55, 129-130; set, 58, 83
Very, with past participle, 43, 102-
103
Viewpoint, 35, 118
Wa’nt, 57, 96
Want for you, 56, 92
4 * *
Wants in, 49, 93
Was
after wish, 37, 89
with you as subject, 56, 91
Wasn’t, after if, 30, 88
Way
adverb, 31, 117
way I do (omitted preposition),
28, 109
ways, 47, 71
When, after is in definitions, 41,
115
Whence after from, 34, 110
Where
after I read, 45, 115
where . . . at, 49, 111
Which
indefinite antecedent, 24, 76
reference of, 76
with personal antecedent, 55, 73
Whip cream, 48, 96
Who as object, 35, 80
Whoever as object, 36, 80
Whom for who, 53, 79
Whomsoever, 118 ~
Whose, gender of antecedent, 27,
73
Why, 28, 111-112
Will vs. shall, 29, 41, 83-84
Wish for, 41, 120
Woods, as singular noun, 40, 70
Works, noun singular, 34, 70
Worse of it, 43, 107
Would vs. should, 29, 84
You as impersonal pronoun, 30, 72
Yourself vs. you, 43, 72
0 )
5 * v ; ‘■w* r. *
■
/ o