ABSTRACT

AN INVESTIGATION OF THE EFFECTS
OF INCORRECT GRAMMAR 0N ATTITUDE AND

COMPREHENSION 0F WRITTEN ENGLISH MESSAGES
by Robert A. Sencer

This study is an experimental investigation of the effects on
readers of grammatical error in written English. The dependent variables
‘were assessment of quality writing, attitude toward the writing, and
comprehension. The study grew out of many years of experience in teach-
ing composition and writing at collegiate undergraduate and graduate
levels and the informal observations over those years of the apparent
fact that grammatical error did not seem to affect comprehension nearly

as much as it affected the attitude of the reader toward the writing.

Language theorists have posited that the English language is
highly redundant in a number of ways. If this is true, then errors in
grammar might be overcome somewhat by redundancy, it was hypothesized.
However, the fact of grammatical error seemed to indicate also that if
the incidence of error was perceived such perception would have some
effect on the reader. Language theorists also had posited that in addi-
tion to comprehension or perceiving the conceptual meaning of written
text, there was the possibility of an attitudinal component to the read-
ing response. It was hypothesized, therefore, that it would be the
attitudinal part of the response that would be affected by the incidence

of grammatical error.

Robert A. Sencer

Practice among collegiate educators and administrators has been to
predict students' success in college partly by an examination of their
verbal aptitudes as measured by the College Board SAT-Verbal. Since
these measures are used also to discriminate among students in terms-of
reading ability, it was hypothesized that different levels of verbal
aptitude would have some effect on the differential responses, if any,

caused by different degrees of grammatical error.

The study included an experimental situation wherein 350 male
college freshmen were given lZOO-word essays to read, followed immedi-
ately by quality assessment, comprehension, and attitude tests. Five
degrees of grammatical error were used, ranging from zero error to a
degree containing some 51 errors within about 500 words. Grammatical
error was defined as that contained commonly in some 32 college-level
composition handbooks of recent publication and as defined in Themes,

Theories, and Therapy (Kitzhaber, 1963). Measures of quality assessment,

of attitude, and of comprehension were made and appropriately analyzed.

The results of the study show (1) that the incidence of error was
perceived by the Ss; (2) that attitude toward the message was affected
by grammatical error; (3) that comprehension appeared not to be affected
by grammatical error; (4) that level of verbal aptitude did have an
effect on the perception of error; (5) that level of verbal aptitude did
not have significant effect on attitude, but that there was some inter-
action between verbal aptitude and attitude; and (6) that verbal aptitude

did not appear to have an effect on comprehension.

Robert A. Sencer

The study suggests that the concept of "incorrectness" as used

I
\ .

commonly by English teachers and critics of writing needs further study

and that the area of response to written messages needs further study.

AN INVESTIGATION OF THE EFFECTS
OF INCORRECT GRAMMAR ON ATTITUDE AND

COMPREHENSION IN WRITTEN ENGLISH MESSAGES

by \

L"
‘L

Robert A:”Sencer

A THESIS

Submitted to
The Department of Communication
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

April 1965

ACKNOWLEDGEMENTS

It is a distinct pleasure to acknowledge herewith the thoughtful,
serious, and kind guidance given by the faculty and staff of the Depart-
‘ment of Communication, Michigan State University. Special acknowledge-
ment is made here to Dr. Erwin Bettinghaus, my committee chairman, and
Dr. David Berlo, chairman of the department, not only for the normal
guidance such positions demand, but also for their willingness to give

guidance over long distances and times. For all of this, I am grateful.

Two men not in the Department of Communication must be mentioned
here: Dr. Henry Lee Smith, Jr., Chairman of the Department of Anthro-
pology and Linguistics, State University of New York at Buffalo, and
Dr. Hans Wolff, Department of African Studies, Northwestern University,
both of whom are responsible for generating a new disciple in the study
of language. Both of these men taught and guided me wisely and consid-

erately. For this I am grateful.

To my own colleagues, students at Michigan State and faculty at
Rensselaer, I must also state my appreciation for their encouragement,

forebearance, and kindnesses.

Most important, my acknowledgement of my family, all of them, is
gratefully made here -- particularly for their encouragement and

patience when understanding was impossible.

ii

.CHAPTER

II

III

IV

BIBLIOGRAPHY

APPENDIX A

APPENDIX B

TABLE OF CONTENTS

PREFACE

THEORETICAL RATIONALE

Rationale
Hypotheses

METHOD
Subjects
Design
Procedure

RESULTS

DISCUSSION

Hypotheses
Implications for the Future

Sample Tests

Composition Handbooks

iii

PAGE

31
31
31
35
39
47

47
54

57
61

107

LIST OF FIGURES

FIGURES PAGE
1. Two-Fold Chi Square Analysis 33
2. Two-Part Two-by-Five Factorial
Analysis of Variance 34
3. Rankings of the Panel on Five
Test Versions 37

LIST OF TABLES

TABLES PAGE
1. Chi Squares -- Well-written 39
2. Chi Squares -- Badly-written 4O
3. Chi Squares -- Totals 41
4. Analysis of Variance,Attitude 42
5. Analysis of Variance,

Comprehension 43
6. Duncan Multiple Range Test

Attitude -- High Verbal 44
7. Duncan Multiple Range Test

Attitude -- Low Verbal 44
8. Means -- Attitude 45

iv

PREFACE

The Project English Research Conference report, Needed Research

 

.ig the Teaching of English (1963), states that there are several problems

which should be studied carefully and soon.

The problem which was mentioned most often in
the papers and discussion groups and on which
the conferees probably spent more discussion
time than any others was the effective use of
language. The first question was whether gram-
mar should be taught at all and, if so, which
grammar: traditional, structural, generative, V-
or some synthesis of all three. The next ques-
tion was the relevance of grammar in one form
or another to the teaching of writing. Research
suggests strongly that there is little correla-
tion between an understanding of grammar and the
ability to write well.

(Needed Research ..., p. 126).

In "Needed Research in the Language Arts," (Research Methods in

the Language Arts, 1961, p. iv) Dr. Carlton M. Singleton says:

 

However, an examination of the teaching of
English in today's schools suggests a need for
more than a compilation of research sugges-
tions. Despite the findings of the past,
today's teacher of English must be shocked when
he realizes how little we know. We are still
unable to describe the teaching-learning pro-
cess accurately; to guide students to a full
and efficient use of their talents; or even to
describe and delimit that which must be taught.
...The true answerEto the problem of knowing
what to teachjis that we know little more today
about the teaching of English than we knew fifty
years ago.

Usage, grammar, and composition are taught now
as they were in 1910 deSpite prettier textbooks
and better trained teachers.

Research, praiseworthy as it is, has demonstra-
ted but little of that which we need to know to
teach English well. If the present rate of
progress is to continue it may well be a

-2-

thousand years hence before much improvement
in our teaching is evident. We can't wait.
We need a miracle.

One of the particular problems in trying to "know" more about how
to teach English is finding out how the various elements which seem to /
constitute "good" English affect the reader. Many studies have been done
in which the scores of the dependent variable have been based on teacher
evaluation: Anderson, 1940; Anderson, 1960; Banargee, 1939; Buck,
Paxton, and Hyndman, 1961; Coward, 1955; Diederich, French, and Carlton,
1961; Foster, 1928; Gerber, 1948; Huddleston, 1952; Kitzhaber, 1963;
Mann, 1944; Norberg, 1951; Perrin, 1933; Powell, 1934; Smith, 1956;
Thompson, 1955; Toporpwski, 1958; ad infinitum. But in the experience
of this writer there have been 22 studies which attempt to assess the

effect of "good" writing, or bad, on what might be considered the "normar'

reader.

It should be noted that only recently, during the Project English
Research Conference mentioned earlier, have English teachers, particu-
larly those interested in research, come to ask questions about how best
to teach students to write better, and about the need for overt and con-*

scious cognitive behavior on the part of the student.

The 1959 report of the Modern Language Association, "The Basic
Issues in the Teaching of English," (PMLA, 1959) indicates quite clearly
that English teachers in the United States are anything but satisfied

with what has been accomplished in this area to date.

-3-

When we proceed to look at the present state
of English in the United States, from the
kindergarten through the graduate school, we
find that the many years of exposure to the
subject and the good and simple reasons for
studying it seldom combine to form a satisfy-
ing picture. Some hostile critics have said
that if as much student time were spent on any
other subject with so little in the way of
results, it would be a national scandal. .
Defenders would reply that English is extremely
broad and general, the results are not easy to
measure [my italics], and the efficacy of
English teaching should not be judged by its
poorest products.

 

(PMLA, 1959) .

CHAPTER I
THEORETICAL RATIONALE

There is need for research into language problems such as the
effects of incorrect grammar on whatever criteria we can muster and for
which we can devise some reasonable methodology for their measurement.
This is not a new idea at all. In 1712 Jonathan Swift published A £52-
posal for Correcting, Improving, and Ascertaining the English Tongue.
Though probably not the first English grammar, Swift's attempt was cer-
tainly one of the very early ones and was followed by a continuing series
of similar attempts to make the language better:/or at least better used,
to make people use it "correctly," and to describe it accurately. This
series, although in no way a formal one, is still continuing, and at

present seems to be enjoying a heightened popularity.

Teachers of English, speech, and all the language arts have
assumed that "correctness" in grammar was a universal perfection to be
achieved by anyone who wanted to be considered "educated," or "a gentle-
man," or "an effective Speaker." The justification for such an assump-
tion seems to have been, and seems to continue to be, a kind of empirical
correlation of success in public life and the manner of writing or speech.
It is true that the great writers and speakers appear to be "correct,"
although Shakespeare is noted for having used, "...between you and me...";
Jonson for substituting whg_for 3222; Ruskin, Chesterfield, and even
Churchill for using singular verbs with plural subjects and vice versa;

a host of both old and modern writers for dangling modifiers in front of

-5-

the eyes of the beholder; more for using this without a sufficient ante-

cedent; and so forth and so forth.

Yet literally every major publisher in this country promotes a
handbook of grammar and other aSpects of good writing, in some cases more
than one such handbook. The vast majority of such books over the past
twenty-five years have been based on the prescriptive grammar of the
eighteenth and nineteenth centuries, a grammar which is quite obviously”
taken from that of classical Latin. Lloyd and Warfel (1956) give an
excellent and brief description of the changes in English from 400 to
1550 showing how much the language has depended upon a Latinate grammar
even though the lexicon is made up of borrowings from a number of lan-
guages, including Anglo-Saxon and Latin. Gorrell and Laird (1953), on
the other hand, reveal the nature of such books in the following excerpts:
" (2) Good writing -- as manifested in word
choice, in sentence structure, and in the
broader aspect of composition -- must be based
on clear thinking.
" (3) Amastering of writing can be acquired
most quickly and thoroughly through improvement
rather than correction; so—called errors dis-
appear only when the writer learns to construct
a sentence well.
" (4) ...The student of English derives the
greatest benefit from the study of English
grammar, not the grammar of another language
-- not even that of one as excellent as Latin
..."(Gorrell and Laird, 1953, 1956, p. iii).

In item (2) the editors seem to shy away from the kernel of the problem

in making the assumption that "clear thinking" lies at the base of what-

ever they mean by "good writing." In (3) they enunciate what appears to

-5-

be a tautology. And in (4) they appear to err by implying that English

grammar is different from Latin grammar.

Among the oldest of the handbooks still in popular use today is

Woolley, Scott, and Bracher's College Handbook 2: Composition. This

 

dates back to the original Woolley edition of 1907. In the preface of

the current edition is this sentence: "The purpose of the College Hand-

 

.2225 still remains that of providing both students and teacher with a
reliable, intelligent, statement of the fundamental principles of
American usage" (Woolley, Scott, and Bracher, 1951, p. v). Implied in
this statement is the notion that usage is based on some conceptual prin-
ciples, a notion that modern linguists all decry most vigorously. The
fact that the handbook does contain a section, a sizable one, on grammar
and syntax is sufficient evidence that the editors believe that grammar
is at least one of the factors forming the "fundamental" basis for the

teaching of written communication.

Kierzek's Macmillan Handbook 2: English contains the following:
"...it is still a rhetoric and handbook combined..." (p. vii), and "The
first part of the book attempts to give the beginner the sort of helpful,
common-sense advice about writing that he needs the most when he is a
beginner...He is then taken through discussions about grammar as a tool of
effective writing..." (p. vii). Again such statements as these quite
obviously show a kind of academic carelessness which seems to be tradi-
tional in the whole school of prescriptive grammarians. First, one must
comment about the reference to "helpful, common-sense advice" which the

author promises the student. Such advice is frequently thought of as

-7-

being the result of some kind of intuitive knowledge, which may be quite
accurate and useful but which in itself shows little evidence of its
accuracy and usefulness. Second, one must comment on the references to
the "beginner." It is quite apparent upon only casual examination that
the book is meant for upper high school and college students. {Thggglggg-
ple are not beginners in any sense of the word. They know their language
and its structure and they have known it since they were five years old!
Third, one must comment on the phrase "...grammar as a tool of effective
writing." This phrase contains, in a sense, the hypothesis for this
entire study. Kierzek postulates the grammar is a tool of effective writ-
ing. One wonders what Kierzek's definition of effective writing is; one
wonders what he means by the concept of tool; one may even wonder what he

means by grammar.

It must be added here that the reference to college and high school
handbooks of writing does not mean that one should arbitrarily question
all that is in them -- far from it. But it is clear that either by impli-
cation or by outright postulation, teachers of English have for years
taught that "good grammar" is a necessity for "good communication" and
have given no evidence for their statements whatsoever. Yet one fre-
quently hears what any teacher of English would call bad grammar and one
can observe that the communication resulting therefrom is not always
"not-good." It is clear that grammatical error in spoken communication
is much more likely to go unnoticed than is grammatical error in written
communication. But one also begins to wonder just £22 is it that does

the noticing: teachers of English? editors and other would-be writers? A

-3-

postulate to the hypotheses of this study is implied here: communication
is not merely the "transfer of information" from the source to the
receiver; it can be indexed only in terms of the behavior of the receiver. /
This behavior is certainly pluralistic and in this study will be identi-
fied as having as important components comprehension, attitude, and recog-

nition of "quality."

In an article in College English (May 1951) entitled "Anything
Goes," Robert Geist makes a point that needs to be made here. Usage and

correctness are rather ambiguous terms and may at times be synonymous and

 

at other times not synonymous. He says:

It is impossible, of course, to find absolute
lines to separate justifiable;improvement of
sentences and diction, the illusions most of

us have about educated speech, and the pedantry
of those who would keep eighteenth century
rules regardless. As we state our objections
to pedantry, we should try to avoid overstate-
ments that invite counterobjections (Geist,
1951, pp. 454-455).

Robert Hall is even a bit stronger in his opinion when he says:

But to return to our basic point: the message
that linguistics has for our society at present
is primarily that one we have used as the title
of this book: Leave Your Langgage Alonel We
put it this way on purpose, to emphasize that
any meddling with our language, by ourselves or
others, in the name of "correctness," of spell-
ing, or of nationalism, is harmful. (Hall, 1950,
p. 248).

In these quotations is the implication that matters of correctness

are ones that need some looking into, not only from the standpoint of

-9-

trying to find out what is and what is not correct, but even more impor-
tantly in trying to find out how people evaluate things as correct and
incorrect and what effect violations of the "rules" or the customs of the
language have on the people who are the receivers of the messages. This
emphasis on the behavior of the receiver should not be interpreted as
meaning that the intent of the source should be ignored. Quite the con-
trary! But, as will be shown later, the source is a receiver, the
receiver a source; and the structure of the language (we are restricting
ourselves here primarily to the verbal language) may have important bear-
ing on what behaviors are elicited by both generation of messages and

reception of them.

The background linguistic system (in other words,
the grammar) of each language is not merely a
reproducing instrmment of voicing ideas but
rather is itself the shaper of ideas, the pro-
gram and guide for the individual's mental activ-
ity, for his analysis of impressions, for his
synthesis of his mental stock in trade. Formu-
lation of ideas is not an independent process,
strictly rational in the old sense, but is part
of a particular grammar and differs, from
slightly to gnntly, as between different gram-
mars. We dissect nature along 1ines laid down
by our native languages. The categories and
types that we isolate from the world of phenom-
ena we do not find there because they stare
every observer in the face; on the contrary, the
world is presented in a kaleidoscopic flux of
impressions which has to be organized by our
minds -- and this means largely by the linguis~
tic systems in our minds. We cut nature up,
organize it into concepts, and ascribe signifi-
cances as we do largely because we are parties
to an agreement to organize it in this way --

an agreement that holds throughout our speech
community and is codified in the patterns of

our language. The agreement is, of course, an
implicit and unstated one, BUT ITS TERMS ARE
ABSOLUTELY OBLIGATORY; we cannot talk at all

-10-

except by subscribing to the organization and
classification of data which the agreement
decrees. (Whorf, 1956, pp. 212 ff).

The foregoing quotation has been included for two reasons: first
to introduce the idea that the grammar of a language is considered by lin-
guists such as Whorf as an element of primary linguistic import, and sec-
ond to begin to define, for the purpose of this study, what grammar means.
It must be stated that Whorf's use of grammar, similar to those of Sapir,
Bloomfield, Fries, and the entire modern school of American linguists, is
most general. The term is used to indicate the system in which elements
of the language go together in an orderly and structured way. Indeed,
for Wharf, the term grammar and the term structure may be considered as
synonymous. For the purpose of the proposed study, however, it is neces-
sary to operationalize the definition and to justify such operationaliza-

tion in terms of current usage among modern linguists.

When one proceeds to lower levels of abstraction, it becomes evi-
dent that there are some disagreements among linguists, or at best, that
there are some areas about which there is no agreement. Gleason, for
instance, defines grammar very simply at one point: "Grammar deals with
the morphemes and their combinations" (Gleason, 1961, p. 11). Gleason
goes on, of course, to define grammar more Specifically, but always
within the framework of this earliest definition. He indicates that
grammar encompasses the study of the relationships between various com-
binations of morphemes (but he seems to mean simply words here) which
have one word common to all the combinations. He includes in his grammar

the study of substitutions of one combination for another and he

-11-
concludes with this overall statement:

The fixed order of morphemes in certain con-
structions and the definable degree of freedom,
are basic to language. They are expressions of
the systematic structure which is the real
essence of Speech. It is the business of lin-
guistic science to describe these principles of
arrangement in the most comprehensive and con-
cise way possible. Such a description is the
grammar of the language. The term is in poor
repute with some, largely because of lack of
precision in its use, and because it has fre-
quently served as a label for legislation as to
how a language should be used, rather than as a
description of how it actually is used...gram-
mar will comprehend two convenient, but not
precisely delimitable, subdivisions: morphology,
the description of the more intimate combina-
tions of morphemes, roughly what are familiarly
called "words"; and syntax, the description of
larger combinations involving as basic units
the combinations described under the morphology
of the language. Some linguists use the term
morphology to cover both subdivisions, in which
case it is equivalent to grammar as used here.
(Gleason, 1961, pp. 57-58).

Nelson Francis has suggested a rather useful description of gram-
mar by defining it into grammars l, 2, and 3. Grammar l he defines as
"the set of formal patterns in which the words of a language are arranged
in order to convey larger meanings." By this he seems to mean the
learned patterns that the native speaker "picks up" as he matures from
the stage of non-conscious maker of noises to the stage of non-self-con-
scious communicator and user of the patterns of the language (at about
age five and a half). Grammar 2 is defined as "the branch of linguistic
science which is concerned with the description, analysis, and formuli-
zation of formal language patterns." It has seemed to some that there is

not sufficient difference between Francis' grammars l and 2 to make a

-12-

difference, but it can be shown that the difference does exist and that
it is a very significant one. In the first place, it is clear that the
five-year-old child does use a set of patterns reliably, over and over
again without significant change. It is also clear that within any
national language culture any given group of five-year-olds uses the same
set of formal patterns -- over and over again without significant change.
Beyond this, when the younger-than-five tries to communicate with his
older brothers and sisters in the culture and uses the right words and is
not successful, his lack of success is attributable to and only to his
confusion of patterns or his lack of pattern, at least from the point of
view of the listener. But the child does not have much of Grammar 2; he
cannot tell us much about the patterns he uses. Grammar 3, as Francis
describes it, is something this study is also concerned with, "linguistic
etiquette." Francis expressed his thinking rather delightfully:

The third sense in which people use the word

"grammar" is "linguistic etiquette." This we

may call "Grammar 3." The word in this sense

is often coupled with a derogatory adjective:

we say that the expression "he ain't here" is

"bad grammar." What we mean is that such an

expression is bad linguistic manners in certain

circles. From the point of view of "Grammar 1"

it is faultless; it conforms just as completely

to the structural patterns of English as does

"he isn't here." The trouble with it is like

the trouble with Prince Hal in Shakespeare's

play -- it is "bad," not in itself, but in the

company it keeps. (Francis, 1954, pp. 299-300).

Fries (1952, p. 56) says, "...the grammar of a language consists

of the devices that signal structural meanings..." and goes on to indi-

cate that such structural meanings are those which are conveyed by

-13-

patterns of arrangement (grammar?) and the selection of form classes
(parts of speech?), as contrasted with lexical meanings, the meanings of
the forms themselves. (This latter meaning, of course, is what Morris
(1958) says is the result of "semantic rules" of the language as opposed

to "syntactic rules.")

In this study, however, grammar will mean that formal pattern or
set of patterns of discourse used by the majority, in fact by all, of the
people in the culture. Deviations or departures from "correct" grammar
will be defined in the design section of this study in some detail, of
course, but for the present it should suffice to state that what is con-
sidered correct in this study will be that pattern or set of patterns

which is common to all formal written public discourse.

Now to a lower level of abstraction and generalization, a survey
by this writer of some thirty-two current handbooks of composition and
grammar (see Appendix B for the list) reveals what may be significant
agreement among the editors of what constitutes grammar at this level.
The rules which were labeled under the rubric grammar include, generally,
the following:

parts of speech

pronoun form (case and agreement)

verb form (tense, mood, principal parts)
subject-verb agreement

adjective-adverb form

conjunctions (coordinating and subordinating)

The last category was the least common of this list, and was included in

-14-

a nmmber of the handbooks under the rubic of word choice. The agreement

 

among teachers of English in this country on what constitutes grammar is
almost amazing when viewed in juxtaposition to the disagreements among
the same teachers of English as to what constitutes the rest of the dis-

cipline called composition instruction.

Kitzhaber's Dartmouth study (1963) tends to sum up the state of
the art at this point so well that it seems needless to go into the jus-
tification of selection of these "rules" of the language any further.

In this study a fairly careful examination of the writing of college stu-

dents was carried out to determine whether instruction in grammar and

writing had any measurable and significant effects on the writing of col-

lege students. The report states that although some effect was noted

between the freshman and SOphomore years, by the end of the senior year
L/

the students were actually writing with more errors than they had when

they were freshmen. Some of the descriptive statistical data in the

Kitzhaber study may be taken as authoritative and useful. Studies were

made of freshman writing, with the "errors" categorized into nine groups,

one of which was labeled grammar. This group contained the following

kinds of errors:

Nouns: error in number

Nouns: error in case

Pronouns: error in case

Verbs: error in principal part

Verbs: error in tense

Verbs: error in mood

Incorrect use of adverb or adverbial element

Incorrect use of adjective or adjectival element

Error in agreement: subject - verb

Error in agreement: pronoun - antecedent
(Kitzhaber, 1963, p. 44).

OCCDNO‘UT-PUONH

H

-15-

Kitzhaber's report does not, unfortunately, further operationalize these
terms.

The next theoretical question to be posed is that of hypothesiz-
ing from the current state of knowledge about our language what effects
should be expected from.what we are calling mistakes in grammar. Trager
and Smith (1957) make it very clear that the lexical item we call a word
may "carry" part of the message, that the inflectional affixes may carry
part of the message, that the phonemic suprasegmentals carry part of the
‘message, that the combinations and orderings (syntax and grammar) carry
part of the message, and that the suprasegmental phrase morphemes carry
part of the message. In fact, the Smith-Trager theory is that all of the
above parts of the message are necessary to the "carrying" efficiency of

the communication process.

Miller (1951, p. 106), in discussing redundancy, says, "In one
form or another, every language, if it is to be reliable, must be redun-
dant. There is no other way to catch mistakes. Thus the time spent
redundantly is not wasted, for it serves to make the language a depend-
able channel for communication." He goes on to indicate that there are
three kinds of rules of language use which are easily distinguishable by
the student of language: semantic rules, syntactic rules, and pragmatic
rules. Miller appears to suggest that a mistake in the use of one kind
of rule may be overcome by the redundancy existing in the use of the
other kinds of rules. What he seems to ignore is that there are built-in
redundancies in each category of rules as he delineates them. But when

he says (p. 103), "Redundancy has its advantages, and a large degree of

-16-

interdependence among the successive units of a language means that parts
of the message can be lost or distorted [incorrect grammar?], without
causing a disruption of communication," it appears that he is taking into

account the built-in redundancies at each level.

Lloyd and Warfel (1956, p. 324) say, "These three ranks of signals
[phonemes, morphemes, and syntactic patterns] strike the ear all at once
as a single impression within a large number of variations in the stream
of sound that we do not hear because we know they do not count; they have
no relevance to the meaning." Later they say (p. 328), "With it [phonemhz
notation used as a tool for observation] we expose the repetitions and
regularities by which we communicate; we break the barriers of sound and
penetrate to some extent the mystery." What is of interest here is that
the redundancies, to use Miller's term, built in to the three "ranks" are
such that they work together to the extent that we are not normally aware
of them and so that they "cover" the "noise" that may exist in the system.
It may seem to be a pertinent question to ask what Lloyd and Warfel mean
by the term repetitions in this context, but it should be clear that they
are not speaking of the kind of redundant repetition that we think of as
so necessary in Spoken language. They are talking here of the regulari-
ties of patterns of phonemes, morphemes, and syntactic constructions,
which regularities suggest to us that language is a system with describ-

able patterns.

Hockett in discussing the design of a language (Hockett, 1956,
p. 301) says, "A language is a complex system of habits." The italics

are his. Here we can suggest a new note in the description of how

-17-

language operates which may indicate another way in which our responses

to language utterances, rather than our generation of linguistic sounds,
have important bearing on this study. If, as Smith says, we "know" the
grammar of our language by the time we enter elementary school, we may
posit that we have acquired a good many, almost the necessary number of
linguistic habits. If we have acquired these habits so thoroughly, then
when we encounter an "error" in linguistic usage, we will tend to "cover"
for it, to get closure, to perceive the stimulus, the message, as a cor-
rect and complete message. Just as the many studies in gestalt psychol-
ogy of visual perception have shown that the human organism seeks after
closure in any complex stimulus situation, so it should be true that in

a linguistic situation the organism will seek closure. Taylor's cloze
procedure (1954) is posited on this same basis, and where Taylor has used
cloze procedure as a measure of the difficulty of reading in a written
message, the subject is forced to "cloze" the gap in the mutilated mes-
sage. Taylor, in fact, defines the subject's responses as the reproduc- \//
tion of the source's linguistic habits in the mind of the receiver.
Hockett makes it very clear that linguistic habits have a place in the
brain when he says, "...but what he does deduce or learn from these obser-
vations is abstracted from the speech and the situations, and established

as a set of patterns, in the brain of the child, in the brain and note-

books of the analyst" (Hockett 1956, p. 302).

Even Jesperson (1921), as incomplete and in some cases erroneous
as he has been shown to be, was cognizant of the habit formation process

in the child.

-13-

To learn a language it is not enough to know
so many words. They must be connected accord-
ing to the particular laws of the particular
language. ...At first each word has only one
form for the child, but he soon discovers that
grown-up people use many forms which resemble
one another in different connexions [Sic], and
he gets a sense of the purport of these forms,
so as to be able to imitate them himself or
even develop similar forms of his own.

The countless grammatical mistakes made by a

child in its early years are a tell-tale proof

of the difficulty which this side of language

[:grammar] presents to him -- especially, of

course, on account of the unsystematic char-

acter of our flexions and the irregularity of

its so-called 'rules' of syntax.

(Jesperson, 1921, p. 128).
This latter paragraph reveals a point of view no longer thought to be
true. Brown and Fraser (1963) have shown that (1) children do have
construction rules, (2) that they do more than just memorize patterns,
that they must somehow assimilate the patterns and then apply them in new
situations, and (3) that child speech is a systematic reduction of adult
speech "largely accomplished by omitting function words that carry little
information." Church (1961) also shows that the child has the rules of
u/

his language at a very early age: "The composing of statements points to
the fact that in learning language the child does not merely acquire a
stock of words. As Brown and Berko [l960]and Ervin [1957] have pointed
out, the child also learns what adults know as the 'rules' of grammar and
syntax -- rules of flexion for tense and mood and number (and, with pro-
nouns, gender and case), of word order, and eventually, of constructing
compound and complex sentences." None of the above mentioned investiga;

./
tors tries to indicate that the child's language is without error, but

-19-

Church points out better than most how the child's errors do occur.—
"When adults skilled in stenography, including those with considerable
experience with children, are asked to record verbatim the speech output
of a preschool child, their records, compared to one carefully tran-
scribed from a tape-recording, are, in fact, reconstructions in adult
English. The child's actual verbalizations are fraught with gaps, hesi-
tations, elisions, strange inversions of word order, repetitions, false
starts, and irrelevancies that appear from nowhere, often completely
changing the sense of what the child is trying to say. Much adult
Speech, of course, shows the same features, if one is alert to note them"
(Church, 1961, p. 271). It hardly needs to be pointed out that Church's

findings do not include errors specifically identified as grammatical.

If these findings are to be believed, and it is not suggested here
that there is any reason to doubt them, one must assume that the habits
of language are well ingrained, that the intervening variable of habit
strength is strong. The habits then become generalized to the extent
that they permit closure over gaps and over so-called errors in grammar.
From such a deduction, then, we must conclude that errors in grammar do
not necessarily interfere with the comprehension of written messages, at
least when we consider errors in grammar to the extent, even, of the
Kitzhaber example quoted earlier. From the theories of learning one may
add another factor to support this conclusion. If the closures over
errors in grammar accomplish what may be in effect the correction of the
error, the resulting comprehension of the message, even within rather

small sequences of discourse (a matter, say of three or four words) may

-20-

act as reinforcement, and may react to the time gradient, which in this
case would be very short, with the result that the reinforcement becomes

a significant kind of reward in the decoding process.

Hypotheses

This study is an elemental investigation of the effects of errors
in grammar on communication. Several hypotheses have been tested, with
several different criterion measures being used to determine what
effect(s), if any, errors in grammar make. These criteria are assess-
ment of writing quality, attitude toward the message, and comprehension

of the message. The specific hypotheses are listed below.

1. In samples of written English text, incidence of grammatical
error will result in the subjects indicating that the text is "badly writ-
ten." In other words, whether attitude and/or comprehension is affected
or not, the subjects will Show that they are aware somehow of differences
in the degree of grammatical error, without their being required to

specify kind or amount of error.

2. In the same samples of written English text, the greater the
degree of grammatical error the greater the differences in the subjects'

attitudes toward the message.

3. In the same samples, increasing incidence of grammatical error
will not result in significant differences in immediate comprehension of

the material in the text.

-21-

4. Differences in verbal aptitude inthe subjects will reflect
significant differences in the subjects' ratings of the text as "well writ-
ten" or "badly written." Subjects above the marginal verbal aptitude level
will show significantly more recognition of the incidence of grammatical

error than subjects at and below the marginal level of verbal aptitude.

I
I

J
In the attitudinal measure, differences in verbal aptitude levels

of the subjects will not Show a significant effect. Subjects with mar-
ginal or lower aptitude ratings will not score significantly lower than

those with verbal aptitude ratings above the marginal value.

In the comprehension measures, subjects with high verbal aptitudes
will reSpond similarly to those with low verbal aptitudes. That is, sub-
jects who have verbal aptitude ratings normally thought of as being
marginal or below will not show significant differences in comprehension

from those subjects whose verbal aptitudes are above marginal.

The following section will discuss the hypotheses in detail,
giving appropriate operational definitions for each and indicating the

dependent variable criterion for each hypothesis.

1. Recognition of Writing Quality: For the purposes of this
study writing quality is defined as the reSponse made by the subjects.
The subjects will be asked to indicate whether they think the text sampleV
is "well written" or "badly written." Since writing quality is a very
complex concept and since college students do not give evidence that they
know how to explain writing quality, the &riterion test for this portion

L/
of the investigation will be a closed-end, two-alternative, single-item test

-22-

The English Composition test of the College Entrance Examinations

Board is designed to reveal the level of knowledge, conscious knowledge,
the student has of his language; part of the examination is in the for-
mal grammar of English. Subjects taking part in this investigation have
a median score on this examination of SgZ, well above the national mean
(theoretically 500). This last comment is included to make sure that it
is clear that, by and large, those taking this achievement examination
consider themselves (or perhaps are so considered by their parents) to
be at or near college entrance levels. According to Dieterich (1961)
this tends to eliminate those who might score lower than 250 on this

test and tends also to skew the scores toward the high end. These data
mean, of course, that when we say that the subjects for the current

study have a median test score of 552 that they are considerably higher
than the "average" high school senior in this country. Such a level of
conscious knowledge of the structure of the language, then, makes tenable
the hypothesis that even though errors in grammar do not seem to affect
comprehension of the material in the message, there is recognition of the
existance of errors therein, a recognition which permits the subject to

announce his opinion as to the quality of the writing as indexed by his

perception of incidence of grammatical error.

A preliminary investigation of the material to be used for the
text sample in this study showed that a small sample (n = 29) of college
freshmen evaluated the text as "pretty well written." This was the
desired goal in choosing the text since the hypothesis requires that room

be left at the negative end of the control group's assessment rather than

-23-

at both ends or at the positive end so that the increase in grammatical
error may be reflected, if the hypothesis is supported, in the assess-

ments by the experimental groups.

2. Effect on Attitude: Attitude will be measured in this study
by use of a semantic differential. It has been hypothesized that as the
degree of grammatical error increases, the attitude of the subjectstoward
the writing will change. Semantic differentials have been used so thor-
oughly and so many times as attitude meaSuring instruments that the
instrument hardly needs justification here. However, it may be useful to
report very briefly on its justification. Osgood, 25.21.,(1958, p. 189)
says, first, "One of the significant by-products of our work in experi-
mental semantics, we believe, has been a new approach and rationale for
attitude measurement." These investigators agree with most attitude
authorities that attitude is an intervening variable, that it mediates
behavior between the perception of the stimulus and the overt reSponse to
that stimulus (some indicate also that attitude mediates the very percep-
tion of the external stimulus), and that attitude is a predisposition to
respond in a certain way. Further, Osgood posits that attitude is dif-
ferent from other mediating variables that predispose the organism to

respond in a certain way and adds, "

...that they predispose toward an
evaluative response” (p. 189). His italics. Throughout the entire area
of attitude research there is an assumption that agrees with Osgood's:
that attitudes have direction, for and against, toward and away, favor-

able and unfavorable. This assumption implies then that such predispo-

sitions are what Osgood calls "bipolar" and that the distance between the

-24-

poles constitutes a continuum. Another assumption is that this continuum
reflects the "intensity" of the attitude (Osgood), the "degree" of the
attitude (Doob), that attitude has at least two dimensions of direction
and strength. Finally Osgood, e£.el., indicate their rationale for iden-
tifying part of their semantic meaning, or semantic space, as attitudinal

as follows.

If attitude is, indeed, some portion of the
internal mediational activity, it is, by infer-
ence from our theoretical model, part of the
semantic structure of an individual, and may be
correspondingly indexed. The factor analyses

of meaning may then provide a basis for extract-
ing this attitudinal component of meaning.

In all of the factor analyses we have done to
date...a factor readily identifiable as evalua-
tive in nature has invariably appeared; usually
it has been the dominant factor... It seems
reasonable to identify attitude, as it is ordi-
narily conceived in both lay and scientific
language, with the evaluative dimension of the
total semantic space, as this is isolated in
the factorization of meaningful judgments.

In terms of the operations of measurement with
the semantic differential, we have defined the
meaning of a concept as its allocation to a
point in the multidimensional semantic space.
We then define attitude toward a concept as the
projection of this point onto the evaluative
dimension of that space. Obviously every point
in semantic space has an evaluative component
...and, therefore, every concept [my italics]
must involve an attitudinal component as part
of its total meaning. (Osgood, e£_el., 1958,
pp. 190-91).

The more Specific justifications for the semantic differential,
in terms of correlations with other measures of attitude, are given fully

in the Osgood reference made earlier. It should be sufficient to indicate

-25-

that the semantic differential has correlated highly and positively with
a number of Thurstone scales, with Guttman scales, with Remmers' scales

and with a modified Bogardus Social Distance Scale.

An immediate problem concerning the measurement of attitude has
been presented in this study: if we use the semantic differential, what
concept is being measured? It was originally hypothesized that attitude
would be measured about the "material" of the text used as experimental
samples. It was further hypothesized that attitude toward the source of
the message would also be measured. In pilot studies, however, it was
found that definition of the "material" of the text was unsatisfactory
in terms of the subject's response. In a small sample (n : 32) of the
same type of subjects as those to be used in the main study, there was
almost no agreement among the subjects as to a listing of "material" con-
tained in the text. The second assumption, about the source of the mes-
sage, was discarded since arbitrary and false declarations of sources had
a Significant effect on the apparent attitude of the subject toward the
source. Source credibility could be designed into this current study,
but this variable will make an appropriate subject for future study since

it is one of considerable complexity.

It was decided, therefore, to limit the concept of the attitude
measurement to that suggested by the title of the text sample. Again the
pilot study mentioned showed that the responses to the title remained con-

stant enough over a two-week period (the test-retest coefficient was .88).

3. Effect on Comprehension: In this study comprehension will be

-26-

measured by Taylor's "Cloze Procedure" (Taylor, 1954). Taylor (p. 3)
defines cloze procedure by indicating that cloze scores Show "...the
degree of correspondence between the language habits used by the source
‘while 'encoding' the message (fitting sequences of language symbols to
the meaning) and the language habits used by the receiver while 'decod6
ing' it (fitting meaning to the mutilated message) and, on the basis of
the meaning perceived, attempting to encode those elements that will make
the message's form whole again." The theoretic basis for cloze procedure
is gestalt psychology of perception which posits that a subject will per-
ceive a broken "circle" of stimuli as a completed or unbroken "circle,"
given that other things are maintained constant. This means, in the con-
text of verbal communication and of cloze procedure, that a subject
receiving a sequence of verbal stimuli, a sequence which is structured
such that it forms a "circle," that is, a completed structure, will per-
ceive that structure as a complete one even if it is broken by having
elements actually missing from it. Koffa (1924, p. 322) says, "So long
as activity is incomplete, every new situation created by it is still to
the animal a transitional situation; whereas when the animal has attained
his goal, he has arrived at a situation which is to him an end-situationJ'
Hilgard (1956) comments on this by indicating that since closed spaces
are more stable than open ones and since such closed spaces are per-
ceived more readily as formed figures, the direction of behavior in the
animal is toward an end-situation which brings closure with it. Hilgard
adds what appears to be a simple statement, but one which is profound for
the purposes of this study: "It is in this manner that rewards influence

learning" (p. 105). Katona (1940) has shown that the pattern of word

-27-

responses corresponds to the prior experiences of the subject so that he
supplies either an individual trace (a particular word, for instance) or
a structural trace (a particular part of Speech or a particular idiomatic
phrase). He further reports that structural traces are more likely to
occur than the individual traces in any given situation. It seems obvi-
ous that Katona would agree that what has been defined herein as grammar
includes what he calls structure. Although there is still some question
generally as to the size of the vocabulary items in.one% repertory it far
exceeds the number of structural combinations used in English (or in any
other language, for that matter). Therefore, it seems apparent that it
is possible and even practical to posit that in such a test as Taylor's

a subject can very well duplicate the structure of the original message
precisely. It is on this basis that cloze procedure is used as a measure
of comprehension not only for the cpperelwmeesage (the meSSage with no
grammatical errors) but also for the experimental messages (those with
varying degrees of grammatical error). The intent of the hypothesis here
is to determine to what extent the subject can reproduce the original,,
unmutilated message, not to determine to what extent the subject can

reproduce the mutilated message itself.

Taylor has defined cloze scores in terms of the "language habits"
of both the encoder and the decoder. Skinner (1953) and Keller and
Schoenfeld (1950) have shown that the basic units of verbal communication,
speech sounds, are emitted and reinforced just as any other units of
behavior are. Accordingly, speech behavior can be explained in terms of

the ordinary principles of operant conditioning, and so may "reading"

~28-

behavior. If we apply the Hullian S-R theory of learning, we can posit
Hull's er (generalized habit strength from related habit) as the inter-
vening variable that may account for the trace tendency of a structural
nature. One may assume, then, that the elicitation of verbal responses
in Taylor's cloze procedure is in complete agreement with both a simple

S-R theory and also a mediation theory.

More specifically, Taylor has defended cloze procedure by estab-
lishing high positive correlations between cloze scores and other mea-
sures. In before-after learning experiments, cloze scores were obtained
along with comprehension scores from independently validated multiple
choice test and AFQT intelligence scores. Cloze correlated with the
before comprehension test .70 and with the efpep .88. Comprehension
scores correlated with AFQT .65 and .70, while cloze and AFQT correlated
.73 and .74. These correlations were significant at the .01 level. From
these data Taylor concluded that "...cloze scores, then, may be an index
of learning, comprehension, intelligence, or message difficulty, and it
is a matter of control that determines what they index in a particular

situation."

4. Effects of Verbal Aptitude on Qeelity Ratings of the Writing:
It is hypothesized that differences in verbal aptitude reflect signifi-
cant differences in the subjects' ratings of the writing as "well" writ-

ten and/or "badly" written. The subjects were asked to evaluate the

fl""'n /’
. . . ‘fV’L/JAM“
messages in a Single-item, two-alternative, closed-end question. The

hypothesis says that (1) among the subjects rated as "high" verbal apti-

tude, there will be significant differences among all the treatment

-29-

groups, but that among the subjecusrated as "low" verbal aptitude, there
‘will be no significant differences among the treatment groups. It is pep
hypothesized, however, that there will be significant differences between
the "high" and "low" groups for each treatment, although it might be
thought that there will be significant difference between the "high" and
"low" groups for the most extreme degree of grammatical error. The justi-
fication for this hypotheses does not, unfortunately, lie in the theory
built from prior empirical research. Instead, it lies with the kind of
intuitive assumptions made by teachers of English, at all levels, for

many, many years.

5. Effect of Verbal Aptitude on Attitude: It has been hypothe-
sized that the effects of grammatical error will be affected by the dife
ferences in verbal aptitude as defined earlier. But it is not intended
that this hypothesis include differences in attitude responses between
groups defined as "high verbal aptitude" and "low verbal aptitude." As
indicated earlier, "low" is defined as below 500, high as 500 and above.
The justification for this hypothesis is given in a study by Chotlos
(1958), which shows that attitude ratings of written samples of English
do not correlate highly with intelligence test scores. Since, in general,
intelligence measures do correlate positively with verbal ability scores,

the assumption underlying this hypothesis is again supported.
6. Effect of Verbal Aptitude on Comprehension: Within the limits
of this investigation, it is hypothesized that the kinds of differences

found in college students on the dimension we call verbal aptitude will

make no difference in comprehension as it is affected by varying degrees

-30-

of grammatical error. Verbal aptitude scores are considered by college
admissions people (Morse, 1963) as a Significant factor in the prediction
of collegiate success. And yet no one has identified in any way the cor-
relation between verbal aptitude scores and reading ability. There have
been a number of studies which attempted to correlate verbal aptitude
scores with I.Q. scores, and these, in general, have revealed positive
and moderately high correlation coefficients. However, the application
of such findings to the decoding process has been nil. If the "theories"
of the linguists are to be believed, such measures of verbal aptitude
must measure the ability of the subject to "re-enclode" his knowledge
.EESEE his language, rather than measuring his "knowing his language."

The logic here is that the "ingraining" of linguistic habits results in
the non-conscious ability to overcome errors in the language during a
decoding process, but does not necessarily permit the subject to "talk
about" such errors. If this deduction is valid, then one must hypothe-
size that even such measures as verbal aptitude scores will not reflect

differential abilities to decode messages which have errors in grammar.

CHAPTER II

METHOD

Subjects: Three hundred and fifty male Ss were used in this
study. The 88 were selected from freshman classes of Rensselaer
Polytechnic Institute and were assigned randomly to this study while
being given alternative exercise work to hide, temporarily, the fact
of an experiment. A single lecture situation encompassing some 350
students was available for this experiment, these students being majors
in engineering, science, architecture, and the social sciences. There
is no evidence that this group reflects any special distribution of in-
telligence (except, of course, as being members of this particular
freshman class suggests that they are typical of good college students),
nor does it suggest any peculiarities of time of day (the class met at
11:00 a.m.), interest or disinterest in verbal communication, willing-
ness or reluctance to be part of an experiment. It should be noted that
the 38, by and large, come from one area of the United States, mostly
from the states of New York, New Jersey, and Pennsylvania; that they
come, by and large, from the upper-middle and lower-upper socio-economic
strata of American society; and that they tend to reflect higher mathe-

matical aptitudes than verbal aptitudes.

Design: The original intent of this study was to determine the
effects of grammatical error on the two dependent variables of attitude
and comprehension. But it appeared necessary to investigate also the
ability of $8 to perceive the differential qualities of the written

material, since without such an investigation the assignment of differ-

-31-

-32-

ences in attitude and comprehension, if they did occur, to the differ-
ences in degree of grammatical error would be only Speculation. There-
fore it was necessary to design the study in three parts: first, the.
investigation of the differential perceptions of writing quality (the
first dependent variable) caused by different degrees of grammatical
error (the independent variable); next the investigation of attitude
differences (another dependent variable) due to the different degrees of
grammatical error; and third, the investigation of the effects on com-

prehension (the third dependent variable) caused by grammatical error.

The first part of the study was designed so that the dependent
variable, perception of writing quality, was indexed by the number of
Ss responding with "well-written" and "badly-written." The first hypoth-
esis predicted that the numbers of Ss responding "well~written" would
vary with the degree of grammatical error. A later hypothesis predicted
that the numbers of Ss responding to each degree would be different not
only because of the differences in grammatical error but also because of
the differences in level of verbal aptitude. It was necessary therefore
to obtain frequency counts of those responding in terms of their respec-
tive positions on the verbal aptitude dimension, as well as in terms of

the different degrees of grammatical error.

For these purposes it was apparent that a chi square analysis was
appropriate. Figure 1 below shows the model of the perception of writing

quality part of the study.

-33-

Figure 1

Two-Fold Chi Square Analysis

Dependent Variables Treatments: Degrees of
Grammatical Error

 

Well Written 9 l g e g g
Total
High VA
Low VA
High-Low x2

Badly Written

Total

High VA

Low VA
High-Low X2

 

In Figure 1 above, Xz's are calculated for the treatments for all
the dependent variables: welldwritten high VA, well-written low VA, and
well~written total; also for badlydwritten high VA, badly~written low VA,
and badly-written total. Chi squares are calculated also for the high-
low comparisons of each treatment for both'welldwritten and badly-written.
This will give an analysis of 24 chi squares, shown later in this paper

under Results.

To investigate the effects, if any, of grammatical error on atti-
tude and comprehension, the second and third dependent variables, the
study was designed so that Ss were given a semantic differential attitude
test and a cloze procedure comprehension test, the scores on each being
the index of their attitudes and comprehensions. The design allowed the
$5 to respond to the various versions of the text immediately following

their reading of the control or experimental versions, so that in effect .

-34-

the two tests, one of attitude and one of comprehension, were separate.
To take into account any interaction between the two tests, the adminis-

tration of them was counterbalanced.

Since the data from the two tests permits analysis of variance,
the analysis was made a two-part factorial design. The two dependent
variables were attitude and comprehension, with verbal aptitude used as
an additional independent variable. The first analysis was of main
treatments, that is, of the effects of degrees of grammatical error.
The analysis was designed so that there were actually two separate anal-
yses, one for attitude and one for comprehension;; The second level of
analysis is along the dimensions indicated by verbal aptitude ratings.
It has been suggested that since college admission is partly a function
of a dichotomous division of such measures (at and above an arbitrary
standard score and below that score) that a similar design be used in

this study.

Figure 2 below shows the analysis of variance model used.

Figure 2

Two-Part Two-by-Five Factorial Analysis of Variance

Dependent Variables Treatments: Degrees of
Grammatical Error

 

0 1 2 3 4

 

High VA n-30 30 30 30 30
Comprehension

Low VA 30 30 30 30 30

High VA 30 30 30 30 30

Attitude

 

-35-

Procedure: The procedure was to present to the 85 samples of
English written discourse which was prepared in five versions: one with
no grammatical errors and four with increasing degrees of grammatical
error. As indicated earlier one of the problems to be solved before
beginning was to select text which college students could read and com-
prehend without undue difficulty and which would be of such a substan-
tive nature that it would fit into the experimental situation without
the SS feeling that the material was in any way outside their normal
academic range of interest. Some eight different selections were tried
out in the experimentor's own classes and from a simple rating system
and informal discussion one was selected as meeting these requirements.

Samples of all the versions actually used are included in the Appendix A.

The next problem to be solved was that of inducing into the text
grammatical error which was realistic and again "non-suspicious." A
panel of five English-teacher colleagues was asked to evaluate each of
the versions by circling grammatical errors they found in the text with-
out prior knowledge of what the text was to be used for. Wherever there
was significant difference among members of the panel about a particular
error, that error was eliminated from the text. There was surprising
agreement among members of the panel as to what constituted error in the

given versions, and only six errors had to be eliminated.

To make the errors appear "realistic," a survey of remedial
student writing done over the past five years was made and showed that
the errors used in the study occurred according to the following list,

the number opposite each kind of error giving its incidence per 1000 words.

-36-

subject-verb agreement: 3.6
case of pronouns: 2.5

pronoun-antecedent agreement: 1.6

mood of verb: 1.2
tense of verb: 1.1
principal parts of verb: 0.9
adverbial element: 0.8
adjectival element: 0.6
number of noun: 0.2
case of noun: 0.0

These figures differ, of course, from those reported by Kitzhaber
(1963, p. 46), but this is due, apparently, to the current survey being
taken from writing done in a remedial writing course. On the basis of
these frequencies, the text versions were prepared, the "0" degree ver-
sion having no errors, the "1" degree version having approximately the
distribution of errors indicated in the list above, the "2" degree ver-
sion having twice as many as the "1" version with the same preportions,
the "3” version having three times as many as "1," and the "4" version
having four times as many. It was necessary to check to see that the
various versions not only had different numbers of errors but that the
panel would recognize the differences. The panel's response to being

asked to rate the versions as to their grammatical "quality" is given

below.

-37-

Figure 3

Rankings of the Panel on Five Test Versions

 

Versions Panel Members

A a 9 2 a
0 Degree 1 1 2 l l
1 Degree 2 3 l 2 2
2 Degree 3 2 3 4 3
3 Degree 4 4 4 3 4
4 Degree 5 5 5 5 5

 

The Kendall Concordance Coefficient 0W) showed that the associa-
tion among these rankings was significant at the .01 level (observed
§_= 222, W Z 0.88). Therefore it was assumed that the five versions
‘were significantly different, not only in the number and distribution
of grammatical errors they contain, but also in the effect they had on

the panel of English teachers.

With the test messages so constructed, booklets were prepared,:«
containing complete written instructions to the Ss for the task to be
performed. The order of the cloze procedure test and the semantic dif-
ferential test was counterbalanced. The question asked for evaluation
of the "quality" of the writing was held constant as the last part of
the test. (Capies of the test booklets are included in the Appendix A.)
The booklets were coded for version identification and were ordered “‘
randomly so that the classroom instructor could distribute them without

inconvenience and without having to make any lengthy announcements. The

~38-

investigator was not present during the actual testing procedure, but a
colleague was present to observe the administration. He did not know
any details of the experiment and was requested not to entertain any

questions from the Ss or from the classroom instructor.

Scoring of the tests was conducted immediately following the test
session and any booklets which were not completed were eliminated from ‘
the study. Since the class used numbered some 350 and only 300 88 were
needed for the investigation, absences on the day of the test and in-
completed booklets did not offer any problem. Actually some 324 students
were present, but only 309 booklets were complete enough to be used.

The extra booklets were eliminated by random selection from the appro-

priate versions.

In the semantic differential attitude test the following scales

were used for actual scoring:

good - bad (evaluative factor loading: .88)

valuable - worthless (.79)

pleasant - unpleasant (.82)

fair - unfair (.83)

honest - dishonest (.85)
Masking scales used were weak-strong (.19), passive-active (.14),
large-small (.06), clear-hazy (.59), and rough-smooth (-.46). These were
taken from Osgood, EE.§£°: 1958, p. 37. All the scales were arranged
randomly, both in sequential order and in right-left positioning of the

scale adjectives.

CHAPTER III

RESULTS

The results of the experiments are presented in two parts: first
as the statistical analyses and second as a series of statements indi-

cating the support or non-support of the various hypotheses.

Tables 1 through 3 give the results of the chi square analyses.

Table 1

Chi Squares -- Wellbwritten Responses

 

Variables Treatment Frequencies X2 df
High VA 20 l7 18 10 6 ) 9.8* 4
)- 9.6* 4

Low VA 4 3 4 7 3 ) 2.6

x2 l6.0** 9.8* 14.7** o 1.0

 

df l 1 1 1 1
* p .05
** p .01

Table 1 shows, first, that among all Ss who responded with
well-written to all five versions of the text there is a signifi-
cant difference between high verbal aptitude Ss and those of low
verbal aptitude. It also is demonstrated that for those versions
of the text with no or little degree of grammatical error (treat-
ment versions 0, 1, and 2) significantly higher numbers of $8 of
high verbal aptitude rated the text as "well-written" than 88 of low
verbal aptitude. It is also demonstrated that there is a signifi-
cant difference among the frequencies of the five versions at the

high VA level, but not among the 1ow VA frequencies. Apparently,

-39-

~40-

then, even among those Ss who perceived all the versions as well
written the number or frequencies pertaining to each version indi-

cate significance of difference among the versions.

Table 2

Chi Squares -- Badly-written Responses

 

Variables Treatment Frequencies X2 df
High VA 1 2 4 8 l7 ) 26.4** 4
)- l4.2** 4
Low VA 5 8 4 5 4 ) 2.0 4
x2 - 3.6* 0.0 0.7 8.0*
df l l l 1 1
* p .05
** p .01

Table 2 tends to show much the same results as Table 1, except
for the expected difference in direction. Significant difference
between high and low verbal aptitudes is again demonstrated. It can
be seen also that at the high VA level the frequencies increase as
the degree of grammatical error increases and that the differences
among high VA frequencies are significant, although there is no sig-

nificant difference demonstrated among low VA frequencies.

Table 3, which follows, gives the results of the analysis of

the responses when verbal aptitude levels are not taken into account.

-41-

Table 3

Chi Squares -- Totals

 

Variables Treatment Frequencies X2 df
_0_ L’ _2_ -3_' _‘*_ _

Well-written 24 20 22 17 9 ) 7.5* 4
)- l9.4** 4

Badly-written 6 10 8 13 21 ) 11.8* 4

x2 10.8* 3.3 6.5* 0.5 4.8*

df l 1 1 1 1

 

* p .05
** p .01

Table 3 Show the analysis of totals of frequencies of two
groups of 38, those responding with "well written" and those re-
sponding with "badly written." It is demonstrated that among those
responding with "well written" there is a significant difference and
in the direction hypothesized, and the same is true for those 88
responding with “badly written." And the difference between the
"well written" frequencies and the "badly written" frequencies is
also significant. However, it must be pointed out that the pattern
of significant differences among the treatment pairs (well and
badly written) is not at all clear. One would expect significant
differences at each end, which expectation is supported, but one
would not expect sigdficant difference in the middle, such a differ-

ence being demonstrated for treatment version 2.

The results shown in Tables 1, 2, and 3 demonstrate general

support for the original hypothesis: that incidence of grammatical

-42-

error will be perceived and indexed by the response of "badly written"
and that there will be some differnce in the responses according to the

verbal aptitude levels of the 83.

Table 4 below Shows the analyses of the scores obtained from the
attitude tests.
Table 4

Analysis of Variance --'Attitude'

 

Sources of d.f. Sums of ‘Mean F
Variation ___ Squares” Sguares‘ .____

Between A 4 3430.63 857.66 110.38**

Between B 1 2.93 2.93 0.37
(cells) (4)

Inter. AxB 4 181.71 45.43 5.83*’

Within 295 2296.62 7.78

Total 299 5911.95

* p .05

** p .01

Table 4 shows the analysis of variance for the scores taken
from the attitude test and shows support for the original hypothesis
that attitude is affected by grammatical error. It also demonstrates
the possibility that levels of verbal aptitude are not affective
differentially within this hypothesis. The significant interaction
result, however, is an interesting addition to the expected results

and will be commented on further after the results of the Duncan

-43-

Multiple Range Test are available.

Table 5 gives the results of the analysis of the compxhension

 

 

scores.
Table 5
Analysis of Variance -- Comprehension
Sources of d.f. Sums of Mean F
Variation‘ Squares Sguares .____
Between A 4 5.06 1.26 0.13
Between B 1 36.77 36.77 3.83
(cells) (4) 58.45
Inter. AxB 4 16.62 4.15 0.43
Within 295 2829.05 9.59
Total 299 2442.04

 

Table 5 does not show any significant differences among
treatments, among levels, or of interaction between treatments and

levels. These results support the original hypothesis.

Homogeneity of variance was checked on the attitude analysis only,
since the comprehension analysis did not Show any significant differences.
The minimax test was used and resulted in an F of 10.37; the F.95 for the
appropriate degrees of freedom (4, 29) was 19 plus. Therefore it is

assumed that no effective heterogeneity of variance was present.

since a significant F was obtained in the attitude study and since

the variances appear homogeneous, it now becomes appropriate to determine

-44-

where among the treatment means the differences lie. The Duncan
Multiple Range Test was used for this purpose and its data are pre-

sented in Tables 6 and 7 below.

Table 6

Duncan Multiple Range Test
Attitude -- High Verbal Aptitude

 

 

Ordered Ts T0 T1 T2. T3 T4, R
Means . ___, __ . . , . , , _.__.
27.46 T0 0 0.76 1.16 6.73 8.13 .586
26.70 T1 0 0.40 5.97 7.37 .611
26.30 T2 0 5.57 6.97 .628
20.73 T3 0 1.40 .640

 

Protection level - .96

Table 7

Duncan Multiple Range Test
Attitude -- Low Verbal Aptitude

 

 

 

0‘49“." T? "'0. T1, ,Tzi, T3 . T4 . R
Means _ _ __ _
27.16 T0 0 0.33 3.36 4.10 7.13 .586
26.83 T1 0 3.03 3.77 6.80 .611
23.80 T; 0 0.74 3.77 .628
23.06 T3 ‘ 0 3.03 .640

 

Protection level = .96

From Table 6 it can be seen that the following differences

appear significant:

-45-

Between treatments 0 and 1, 2, 3, and 4;
Between treatments 1 and 3 and 4;
Between treatments 2 and 3 and 4;
Between treatments 3 and 4.

No difference appears to exist between treatments 1 and 2.

From Table 7 it can be seen that the following differences
appear significant:
Between treatments 0 and 2, 3, and 4;
Between treatments 1 and 2, 3, and 4;
Between treatments 2 and 3 and 4;
Between treatments 3 and 4.

No difference appears to exist between treatments 0 and 1.

The significant interaction demonstrated in Table 4 suggests
that although there appears to be no significant difference in the
attitudinal responses of the Ss according to their respective verbal
aptitude levels throughout the treatments, there may be some significant
difference(s) between levels in some of the treatments. Table 8 below

shows the means for all cells and may reveal wherein such differences lie.

Table 8

Means*-- Attitude

 

Treatments
Levels T0. T1 T2 T3 , T4

 

 

 

HVA 27.56 26.70 26.30 20.73 19.33

LVA 27.16 26.83 23.80 23.06 20.03

 

* based on five semantic differential
scales as indicated on page 38.

-46-

Since the Duncan Multiple Range Test shows, for instance, that the
minimum significant difference among all the means is 0.586 and that
the maximum is 0.640 and since the difference between HVA-T3 and
LVA-T4 is 0.70, it would appear that there is a significant difference
between these two means. Similarly, it would appear that thereis not
such a significant difference between, for instance, HVA and LVA To
and HVA and LVA T1. Therefore it might be concluded that the inter-
action among treatments and levels suggests that those Ss of higher
verbal aptitude are more susceptible to attitude change because of

greater grammatical error than are those of lower verbal aptitude,

within the limits of the current study of course.

CHAPTER IV
DISCUSSION

Earlier in this study it was said that English teachers, particu-
larly teachers of composition, had been insistent that correctness in
grammar, as well as correctness in other areas of composition, was a
necessity for good communication. At this point it might be well to indi-
cate that there is another side to the coin. AS a long-time teacher of
composition, at the freshman level and at the graduate level, this investi-
gator has heard over and over again the student query, "But what differ-
ence does it make?" and "But they seem to understand me well enough."
Neither the teachers nor the students have discriminated sufficiently, it
seems, between the receiver's understanding or comprehension of what is
being said or written and his evaluative response to it. This study pur-
ports to make that distinction clearly and precisely and to investigate“

the differential effects of grammatical error in these two areas.

Following is discussion of each of the hypotheses tested in this
study. Each hypothesis will be discussed individually, but a summary dis-
cussion will precede the final statement of implications of the outcome of

this research.

Hypothesis 1: Increasing degrees of grammatical error will result

 

in increasing preportions of the Ss indicating that the text is "badly"
written. The fact that this hypothesis was supported indicates that 83

were cognizant of grammatical error, whether they could "spot" the errors“

-47-

-43-

or not.‘ The preliminary investigation of what the Ss used in this exper-
iment thought of the text indicated that the control message was, in their
‘words, "pretty well written." This pre-study response indicates that the
absence of grammatical error was noted by the $3; the support of the
hypothesis within the study indicates that the presence of such error was
noted by the $8. The support of this hypothesis also indicates thaqiany
change in attitude and/or comprehension as the degree of grammatical error
changes may be considered due to the change in grammatical errorlj But
beyond this it may be assumed from this investigation that although the
patterns of linguistic structure, including grammar, are deeply ingrained
by the time of college years, they are not so automatically or non-cone..
sciously perceived that SS of the type used in this study do notreact to
their appropriate or inappropriate patterning. This conclusion also indi-
cates that the redundancies of the English language are such that they do
not completely obscure such "gaps" as errors in grammar; that is, the
errors do not go completely unnoticed. This conclusion, of course, is

the one hypothesized originally, and was supported in all its aSpects

within the framework of this study.

Hypothesis 2: Increasing degrees of grammatical error will result
in significant differences in the S's attitudes toward the message, and
the differences in attitudinal response will be increasipgly negative as
the degree of grammatical error increases. This hypothesis was supported,“
and we may then conclude that the meaning that a message elicits in the
receiver has, as Osgood has indicated, an evaluative component which we

may call attitudeC/Osgood's studies and those of Tucker (1955) and

-49-

Darnell (1960) have supported the hypothesis that meaning has an attitude
inal component when applied to messages that are more than the concepts
which form most of Osgood's messages in the development of his semantic
differential. None of these studies, however, has attempted to measure
the attitudinal response in terms of comparison with comprehension. If
hypothesis 3 is also supported, the evidence will then indicate that the
attitudinal component of meaning is separable from other components and.
will permit us then to proceed to try to discover what factors in verbal
communication other than grammar affect this component. If, however,
hypothesis 3 is not supported, we will conclude that (a) errors in gram-
mar cause a decrease in comprehension which is itself caused by the atti-
tudinal response or (b) errors in grammar cause a decrease in comprehen-
sion which causes a decrease in attitudinal response or (c) errors in
grammar cause an interaction which cannot be separated by the methods of
this study into its significant components. If we consider (a) above,

we must conclude also that the redundancy built into our language is not
such that it maintains the same attitudinal meaning for the message whether
there is or is not grammatical error present. Conclusions (b) and (c)
above say somewhat the same thing: that the built-in redundancy of English
is not Shch that it covers the attitudinal component of the meaning of

the message.

Hypothesis 3: In written samples of normal English, increasing
incidence of grammatical error will not result in significant differences
in immediate comprehension of the material. Since this hypothesis was

supported, several conclusions may be made. Among the most important is

-50...

that language, particularly the English language, is indeed redundant in
terms of its ability to communicate the "material" or "information” of
the message. In the theoretical rationale for this study it was suggested
that "the fixed order of morphemes in certain constructions and the defin-
able degree of freedom, are basic to language" (Gleason, 1961, p. 57).
If Gleason meant to indicate that the structure of grammar of the lan-
guage was a function of the ordering of the morphemes, and that that
ordering was inviolate, the current study indicates that<idisordering" of
parts of the message does not have a significant effect on the reception)
or decoding of the message. But it should be obvious that so-called errors
in grammar are tantamount to a kind of disordering and that the mainte-
nance of decoding ability in spite of such error must be accounted for by
something which Morris, Carroll, and others have called redundancy. Eng-
lish, in contrast to languages Such as Chinese, Latin, and some of the
American Indian languages, relies heavily on word order for part of its
meaning-eliciting potential. The maintenance of appropriate word ordero/
must be a part of this redundancy and must help "cover" the errors in
grammar which were used in this current study. Indeed, most of the errors
in the current study still keep the basic root of the appropriate word and
only one of the morphemes, usually an affix, accounts for the "bad" gram-
mar. The support of this hypothesis also suggests that noise in the com-
munication system can be tolerated, in terms of comprehension at least,
to a quite extensive degree. It is also a fact that many languages (other
than English and most of the Indo-European languages) do not rely on

/

inflectional morphemes to the degree that English does. 'We have assumed

in the past that these inflectional morphemes, particularly those Showing

-51-

number, tense, case, mood, were necessary for effective communication.

The present study indicates that such is not necessarily the case.‘ Most
of the examples of "bad" grammar in the test samples were bad in that the
inflectional morphemes were not the "correct" ones. Since the study shows
that comprehension under all five degrees of grammatical error was not
significantly different, one must conclude that these inflectional mor-
phemes do not carry the burden of meaning that has been supposed hereto-.

fore.

Hypothesis 4: Differences in verbal aptitude will reflect signi-

 

ficant differences in the Ss' ratings of the text as "well written" and

 

"badly written." This hypothesis was supported and suggests rather
strongly thativerbal aptitude scores have something to say about the abil-
ity of the Ss to discover, in the normal reading process, grammatical
errors and to react to them so that there may be a distinction between
the evaluative response toward the writing at various levels of encoding
abilityi~ That is, Ss who have the higher level of verbal aptitude will
be more cognizant of the incidence of error than will those at the lower
level.l Since the construction of verbal aptitude tests is based on a
theoretical rationale which includes statements such as 83 who can "Spot"
grammatical errors better than other 83 will score higher, the previous
conclusion is obvious. It is interesting to note, however, that the
College Boards Verbal Aptitude Test, until just recently, has not relied
very heavily on a critical reading exercise. Many English teachere have
noted, and sometimes complained, that the English Composition Test, on

the other hand, did rely heavily on the ability of the student to "spot"

-52-

errors and be critical of them; reference is made here to the so-called
"Interlinear," which required the student to read a sample of English in
which there were about the same number of errors as are included in the
Treatment 4 version of this experiment's test samples and to mark them
and correct them. It must be added, however, that not all of the errors
by any means were what has been defined here as grammatical; perhaps only
a quarter of the errors were similar to those included in the current
test samples. Nevertheless, colleges which require the College Board
Examinations rely more heavily on the Verbal Aptitude Examination than on
the English Composition Test, without considerations of the kind that are

represented here in these conclusions.

If

x The prediction that differences in verbal aptitude would not show
significant differences in attitudinal responses was supported, and
indicates that the range of verbal aptitudes represented in this study
were such thatthe 83 at the lower level of verbal aptitude were as
much affected in their attitudinal reSponses as those at the upper level.
The implication here is that the evaluative response of the Ss to errors
in grammar is not necessarily a function of the 88' ability to score
highly on a verbal aptitude test, unless, of course, his verbal aptitude
score is far below those accepted for college entrance. Although it may
seem that support of this prediction would indicate that the evaluative
component of the meaning of a message is the same regardless of the abil-
ity of the receiver of a message to decode it, the limitations of the
current experiment suggest at least that there may be a possibility that

the built-in redundancies of English are sufficient to overcome the range

 

-53-

of differences in ability to encode and decode as represented in this

study.

The additional prediction that differences in verbal aptitude would
not have a significant effect on the comprehension of the message was also
supported, but this result should be viewed as very limited in terms of
the definition of verbal aptitude used in this study. First, it was
defined that verbal aptitude would have only two values: at and above 552,
and below 552. One fact about the range of verbal aptitude for the SS
used in this experiment is that no S was below 400; this fact rules out
people who really have a limited ability to encode and decodel) The par-
tition of the Ss into two groups was designed in parallel to the admission
requirements usually used in determining a person's ability to pursue
successfully a collegiate career in an institution like Rensselaer. Many
colleges, particularly the few remaining state institutions which must
accept epy student who has graduated from an accredited high school in the
given state, accept students whose verbal aptitude score is considerably
below 400. It might be concluded that this prediction was supported
mainly because of the limitation on the verbal aptitudes of the Ss used
in the experiment. 0n the other hand, even those schools which do not
accept students whose verbal aptitude score is below 400 or even 450 still
require the student to be instructed in English composition and to pay
attention to matters such as grammar. This policy, it must be assumed, is
based on some kind of recognition that correctness in grammar is a neces-
sary factor, or in the students' terms a necessary evil, in the educational

process.

 

-54-

A further conclusion from the support of this prediction is that
(within the given range of verbal aptitude scores used in this study SS
have acquired such habits of language, both encoding and decoding, that
they tend to perceive strings of words and/or morphemes in such a way
that errors in grammar have no significant effect and that the two ranges
in verbal aptitude are not significant in terms of discriminating among

SUCh SS 0

In summary then, it can be indicated that the current study has
shown that there is the possibility and even high probability of a sig-
nificant relation between errors in grammar in written English text and
the perceptions of college freshmen.) Readers were aware of some differ-
ence among the treatment versions and since the only difference that has
been induced into the five versions is that of grammatical error, all
other variables having been controlled and/or taken into account, one
must conclude that what they are aware of is the effect of grammatical
error. The attitude and comprehension test results indicate that it is
attitude and not comprehension that is affected by the incidence of gram-

matical error.

Implications for the Future: Since the hypotheses of this study

are supported, a number of future investigations are immediately Sugges-

/
ted. One may well ask the "big" question: what constitutes incorrectness?

 

If, as was quoted from Francis earlier, correct grammar is a matter of
custom as well as structure of the language, how is the custom developed,

in what social and other contexts does custom differ and how does this

-55-

kind of social difference affect the language itself and the reSponse
patterns to the language? One area in the study of communication that
appears to be almost void of precise and useful experimentation is that
of personality.' One may very well ask questions regarding the effect of
so-called incorrectness in grammar or syntax on different personality V,
typologies. One is tempted to suggest even that those teachers of compo-
sition who seem most adamant in matters of correctness may themselves
form a personality typology similar to Rokeach's theory about isolated
persons in a particular belief system (Rokeach, 1960, p. 398). For those
interested in the cross-cultural aspects of communication, an interesting
series of questions regarding the status of grammatical correctness in
native languages and in borrowed languages may be asked. Here it seems
new ideas for investigation are legion. In the area of applications of
information theory to the communication process one may ask many ques-
tions about the "noise" effects resulting from incorrectness, about the
abilities of channels of various kinds to detect and correct incorrect-v
nesses that matter. One may even be tempted to suggest that for ultra-
long-distance telecommunication (so-called space-talk) a computer-like
language analyzer may be constructed to receive communications, evaluate
them in terms of structural correctness (have concurrent messages been
garbled or have elements of the message been left out, and similar ques-
tions), feed back to the origin any corrective measure indicators, re-
receive the corrected message and upon evaluation of it as being AOK,
send it on to the receiver. Such a science-fiction idea will be dependent
on some kind of theoretical knowledge as to the effects of grammatical

incorrectness on the receiver of the message.-

-56-

Perhaps it will be appropriate to end this discussion with the
statement that it appears that so little is known about the experiential
effects of garbling of the syntax of English, or of any national lan-
guage for that matter, that it behooves us to spend a good deal of effort
in the immediate future in building some kind of theory, some kind of

processual theory for the structure of language.

BIBLIOGRAPHY

Anderson, C. C. The new STEP essay test as a measure of composition
ability. Educ. &Ap§ychol. measurements, XX (Spring, 1960),
96-102.

Anderson, Harold A., and Arthur E. Traxler. The reliability of the
reading of an English essay test -- a second study. School
review, XLVII (September, 1940), 521-530.

Barnerjee, Hemshandra. Reliability of repeated grading of English
essays. Indian journal of psycholegy, XIV (January, 1939),
43-500 '

Basic issues in the teaching of English, the. PMLA, LXXIV, No. 4,
part 2 (September, 1959), 1-12.

Berlo, David K. The process of communication.' New York: Holt, Rine-
hart, and Winston, 1960.

Brown, Roger. Words and thinge. Glencoe, Illinois: The Free Press,
' a ‘1958.

Brown, Roger, and Jean Berko. Word association and the acquisition
of grammar. Child development, XXXI (1960), l-14.

Brown, Roger, and Colin Fraser. The acquisition of syntax. In Verbal
behavior and learning. New York: McGraw-Hill, 1963.

Buck, Dorothy 8., Evelyn J. Paxton, and Roger Hyndman. Report on
readipgeassistants for high school English teachers: 1960-
1961. Los Angéles: Division of Secondary Education, Los

“ Angeles City School Districts, 1961.

Coward, Ann F.” A comparison of two methods of grading English compo-
sitions. Journal of education research, XLVI (October, 1952),
81-03.

Chall, Jeanne S. Readability. Columbus: Ohio State University, 1958.
Chotlos, John. A statistical and comparative analysis of individual
written language samples. An unpublished doctoral disserta-

tion, State UniverSSty of Iowa, 1958.

/
Church, Joseph. Lapguage and the discovery of reality. New York:
' Random House, 1961.

Darnell,'Donald D} The relation between sentence order and the com-

prehension of written English. Unpublished mastergs thesis,
Michigan State University, 1960.

-5}-

-58—

Diederich, Paul B. The English composition test and intelligence.
ETS Developments. Princeton: Educational Testing Service,
Spring, 1961.

Diederich, Paul B., John W. French, and Sydell T. Carlton. Factors
in judgments of writing ability. Research bulletin RB-61-15.
Princeton: Educational Testing Service, 1961.

Ervin, S.M. Grammar and classification. A paper given before the
American Psychological Association, 1957.

Flesch, Rudolf. Marks of readabfle style. An unpublished Ph.D.
dissertation, Teachers College, Columbia University, 1943.

Foster, Richard R. Some aspects of teaching and testing in college
freshman English. Unpublished Ph.D. dissertation, University
of Iowa, 1928.

Francis, Nelson. The structure of American English. New York: Ronald
Press, 1958.

Fries, Charles C. The structure of Epglish. New York: Harcourt,
Brace, 1952.

Geist, Robert J. Anything goes. College English, XII (May, 1951),
454-455 0

Gerber, John C. Testing and evaluation in the skillsLof communication.
College English, IX (April, 1948), 375-384.

I Gleason, H.A. An introduction to descriptive linguistics. New York:
Holt, Rinehart, and Winston, 1955. '
Gorrell, Robert M., and Charlton Laird. Modern Epglish handbook.
Englewood Cliffs, New Jersey: Prentice-Hall, 1956.

liall, Robert A. Leave your language alone! Ithaca, New York:
Linguistics, 1950.

liilgard, Ernest R. Theories of leanning. New York: Appleton, Century,
- Crofts, 1956. ‘

liockett, Charles F. A course in modern linguistics. New York: Macmillan,
1956.

liuddleston, Edith M. Measurement of writing ability at the college
entrance level: objective vs. subjective testing techniques.
.Unpublished Ph.D. dissertation, New York University, 1952.
.Jespersen, Otto. Lapguage. New York: Macmillan, 1921.

ltatona, F. Organizipg and memorizigg. New York: Columbia University
Press, 1940.

-59-

Keller, F.S.& W.N. Schoenfeld, Principles of psycholqu, New York:
Appleton, Century, Crofts, 1950.

Kierzak, John M. The Macmillan handbook of Engiish. New York: Mac-
millan, 1945.

Kitzhaber, Albert R. Themes, theories,eand therapy: the teaching of
writing in college. New York: McGraw-Hill, 1963.

Koffka, Kurt. The growth of the.mind. London: Kegan Paul, Trench,
Trubner, 1940.

Lewin, Kurt. Principles of topological psychology. New York: McGraw-
Hill, 1936.

Lloyd, Donald J., and Harry R. Warfel. American English in its cul-
tural settipg. Englewood Cliffs, New Jersey: Prentice-Hall,
1956.

Mann, Mary B. A study of the problem of quantitative differentiation
* of samples of written English. Unpublished Ph.D. dissertation,
University of Iowa, 1944.

Miller, George A. Language and communication. New York: McGraw-Hill,
1951.

Morris, Charles. Signs, language, and behavior. Englewood Cliffs,
New Jersey: Prentice-Hall, 1948. .

Morse, Jack B. Entrance to college: a statistical problem. School and
”. society, LXVI (November, 1963), 294-296.

Mowrer, O.H. Learnipg theory and the symbolic process. New York:
' Wiley, 1960.

Needed research in the teachinLof English. Office of Education, U.S.
Department of Health, Education, and Welfare. Washington:
U.S. Government Printing Office, 1963.

liorberg, H. Orville. Research and the teaching of written expression.
California journal of education research, II (March, 1951),
89-91.

()sgood, Charles, George J. Suci, and Percy H. Tannenbaums The measure-
ment of meaning. Urbana: Univeréity of Illinois Press, 1958.

IPerrin, Porter G. Commensurate grading of English compositions. School
and society, XXXVII (March, 1933), 386-387.

Powell, Raymond L. Valid testing and diagnosis in the mechanics of :~
ninth grade English.composition. University of Iowa Studies
in education, Ix (January, 1934), 95-130.

-50-

Research methods in the hanguage arts. National Council of Teachers
of English, Champaigne, Illinois, 1961.

Rokeach, Milton. The open and closed mind. New York: Basic Books, 1960.

 

Skinner, E.F. Science and the human behavior. New York: Macmillan, 1953.

'Smith, Henry Lee Jr. Linguistic science and the teaching of English.
Cambridge: Harvard University Press, 1956.

Smith, Henry Lee Jr. Syntax and semology. Languages and linguistics,
No. 16 (Georgetown University Institute of Languages and
Linguistics), 1963.

Taylor, W.L. Application of 'cloze' procedure and entropy measures to
the study of contextual constraint in samples of continuous
prose. Unpublished doctoral dissertation, University of Illinois,
1954.

\Thompson, Wayne N. A study of the grading practices of thirty-one
instructons in freshman English. Journal of educational re-
search, II (September, 1955), 65-68. '

Thorndike, Edward L. Teacher's word book. New York: Bureau of Publica-
tions, Teachers College, Columbia University, 1931.

Toporowski, Dolores S. A study of the opinions and actions of elementary
school teachers and pupils regarding the correcting of papers.
Unpublished M. A. thesis, Boston University, 1958.

Trager, George L., and Henry Lee Smith, Jr. An outline of English strug;
ture. Washington: American Council of learned Societies, 1957.

Tucker, W. T. Experiments in aesthetic communications. An unpublished
Ph.D. dissertation, University of Illinois, 1955.

Wharf, Benjamin Lee. Langugge, thought, and reality. Cambridge and
New York: Technology Press and John Wiley, 1956.

Wooley, Edwin 0., Franklin W. Scott, and Frederick Bracher. College
- handbook of composition. Boston: Heath, 1951.

APPENDIX A

SAMPLE TESTS, VERSIONS 0, l, 2, 3, 4

Note: The sample tests included herein are Xerox copies of the original
spirit duplicated test booklets. The paper used in these copies is

Eaton's "Eterna" archives bond, whose long-lasting characteristic accounts
for the slightly grey color.

-61-

Tm

:egree

11:?

MC

1111

l 391

am

to u!

uni

)0 WT WRITE IN THESE BLANI’QS FILL IN THESE BLANKS Version 0

0
Degree Version

 

 

Student' 8 Néme

V. A.

 

 

Class and Maj o:—

Immanuel-7e

 

This is s special exercise which you are required to do during this class
period. The objective is to measure the group's ability to read text of the
kind that is) in your textbooks and to evaluate your ability to respond to it.
Although you will not be "graded" on the work you do during this exercise, a
record will be kept of how well you do and some judgment will be made of your
ability. It will be prudent, therefore, for you to do as well as you can.

First you will be asked to read an essay of some lzsolwords in length.
Read it carefully but do not try to memorize what you are reading. Read at
a speed which seems sppropriste to you in terms of the text itself and your
own reading ability. Speed of reading is not a matter of interest in this
exercise.

Next you will be given s new kind of test to see what you have been able
to' understand from‘the text. When you get to the rest, read its instructions
csrefully; they ere very brief but important. . ‘

Third: and lest, you'will be asked to answer one question about the text,
your answer being one check mark in en eppmpriste Space. Again do precisely
what the instructions given there say.

As you go through the pages of this booklet, 29 NOT TURN BACK 1Q Afl Egg-
epas 292s- ‘

00 nor ASK AN! qussnoss or THE msrnpcroas on or roux Eamon STUDENTS .
None of them,‘probeb1y, knows the answer. Read your instructions carefully
and you will know all you need to for this exercise.

TURN THE PAGE so: AND MED. Reed the essay and continue on through all

the peges of this booklet. When you are finished, hold up your hand.

/.

THE 1mm OF STRATEGY *

It is all too clear at this moment that there are many
ways for a book to begin; and most of those in plain eight
are transparently bad. We are tantalized by the thought that
somewhere among them 93.1 lie hidden a few having such noble
qualities as these: The readers are informed -- perhaps with-
out suspecting it, though in the clearest prose -- 'of what the
writer intends to discuss; yet at the same time, it sounds like
the Lorelei calling. Whereupon these readers resolve into two
groups:” The first, a large and happy family really, will stick
to the hook to the end, even though unimaginad adversities impend.
Further, this group will always think and speak kindly of it, .
and will doubtless have at least one copy in every room. The
second group is most briefly described by stating that it
differs from the first; but the book acts immediately as a sopo-
rific on all unpleasant passions, so, as it is sleepily laid

aside, the sole lasting impression is that “of a good gift sug-

_ gestion.

If we could devise an opening strategy such as that, it
would wonderfully exemplify the theme and aims of the book,
for. 335 concern. throughout gill 333 gig; a M go_r_ selecting
hgggstrategies, even in contexts where the word 'strategy' it--'

self may not be in comics use;

The contexts of interest to us are those in which people

' are at crossépurposes: in short, conflict situations. The

problem of how to begin this book is recognizably of that type,

for certainly you and the writer are at cross-purposes, as our

9: from Williams, J. D. The Compleat strategist. New York:
McGraw-Hill, 1954, pp. 144. '

 

interests are opposed -- in a polite way, of course, but
definitely opposed. For we hope to cases you into a very
difficult type of intellectual activity, while you a rea-
sonable person with enough troubles already, may crave only
raisin-tion or satisfaction of curiosity. This conflict of

interests is essential in the situations we shall study.

Another element is also essential and it is present
here too: Each of us can exert 2932 control over the situa-
tion. tinny ways will occur to you: for one, you may throw
the book at the cat, thus irritating both the writer and the
cat, .‘but at some cost in property, perhaps some in self-
respect, and undoubtedly some in deteriorated relations with
the cat. or you may skim the hard parts, and so on. There
are aspects within the control of the writer, too, such as
the choice and treatment of content -- but it is not necess-
ary to labor the point. And a further characteristic element

,, appears: Some aspects of the situation are not within the
- _ control of either of us; for example, a. multitude of events
in our pasts and extraneous influences during the writing
and reading periods will: play important roles. Of course
this particular problem, of beginning the book in a really
optimum way; has a further characteristic which we shall
henceforth shun, namely, it is too hard -‘- else we 'ahgum

have solved it .

The restrictions on the subject matter being so few and
mild, it follows that the set of conflict situations we are

willing to consider is most notable for its catholicity.

There is no objection, in principle, to considering an
H-homb contest between Kara and Earth, or a love affair

of the'Bsrrctt-Brm-ming type. The contest may he econom-
ic in character, or it may he Musical Chairs. Or it may

be almost any one of the myriad activities which take place
during conventional war. It doesn't follow that we have

a nostrum’for strategic ills in all these fields, but there
is a possibility that our offering any as a method, perform

useful service in any of them.

I ~The method which "will be presented is identified by
the catch phrase gage. M or, time permitting, the 233293
2§\§321§2 a; Strategy. If this is your first encounter with
that unlikely sequence of nouns, the sole reaction is prob-
ably: Why? Well, the idea takes its name from the circus-
stance that. the study of games is a useful and usable start-
ing point.‘ in the study of strateg . That does not really
help, for again we hear: Why? Well, because games contain
many of the ingredients common to all conflicts, and they
are relatively amenable to description and to study. (In-
cidentally, having used the word 'game' to name the theory,

--we then call any conflict a game when we are considering it

in the light of the theory.) /."

/

/

To illustrate the point, let us run our minds over a
Poker game, keeping watch for items which are significant
in, say, a military conflict.- You and four others are thus

studying human nature, under a cyst“ of rewards, you hope.

We note at once that the players have Opposing inter-
ests; each‘wsnts to win and, because the winnings of

one are necessarily the losses of another, their inter-
ests are opposed. This provides the basis of conflict.
‘Wo observe that some elements of the action, being per-
sonal choices, are completely'within your control. And
the some being true for each player, there are elements
which are not within your co.trol; worse, they are con-
trolled by minds having interests inimicsl to yours.
Finally, there are elements of the game that ere not,

.\ under the rules, within the control of any player, such
as the order of the cards in the deck. These elements
may be thought of as being controlled by nature~- who has
a massively stable personality, o somewhat pucklsb atti-
tude toward your important affairs, but who bears you no
conscious malice. These are all surely familiar aspects

of any conflict situation.

"_Another-charecteristic is that the state of informa-
tlon -- intelligence, in the military sense -- 18'3 factor,
and, as usual, is an imperfect and hence troublesome factor;
we don‘t knOW“whst the other fellow's hole card is. There
is also the‘bluff by which you, or the opposition, give false
eVldence regerdlng intentions or strength of forces.'/Other

similarities‘will occur to you; peeple even get killed,

'occasionolly.

"

But the analogy should not he pushed too far. You can

think of many as e etc of warfare vhf. ch are not reflected
in Poker. One tam; will sometimes kill two tanks, in a
thou dot-7. 12; whereas a pair of Jacks alt- ays wins over an Ace- '
high hand in the chowder-m. Of course Poker could be modi-
fied to mate it contain shop down possibilities of this kind,
say hy ruling that an Ace is upei‘ior to any pair, up to
Jacks, whenever sny‘oody's wife phones during the play of a
hand. But the fac' is that games don't or hibit all the
complexities of warfare and of other real-life conflict
situations -- which is precisely why they are usable start-
ing points for a study of strategy. In the early stages
of. developing a theory it just .is not possible simultaneously

to. handle very many interactim; factors.

It is probably clear, then, that games do contain some
the basic elements that are present in almost any inter-
esting conflict situation. Does it follow that we can learn
useful things by beginning a study with them? Not necessar-
ily... It may'be that military, economic, and social situations
are just basically too complicated to be approached through
game concepts. This possibility gains credence from the fact
that the body of Game doctrine now in existence is not even
ahle to cope with full-blown real games; rather, we are re-

stricted at present to very simple real games, and to watered-

dovn versions of complicated ones, such as Poker.

~ " .-' "o .‘7_ "“;1I‘ ..""IF’ 'D“ -
‘ LAJJ j“ ‘J'ILI ‘ "eh, .LJI 1 ALI—.4 A Filtl. ‘JJ'JYJ‘q's:

iii” THE WORD
‘ j " - — 1' ‘ "l "‘ - . «gm -rx ...,,
‘53.} CC...3.LL):.1\ ..L‘S'.‘ 17?; ‘ Hf) 711333,. 7.1.11 In}: To LEAVE Ala? Obi. :"C ATOM; ‘ ivLY

q up-‘f-r- " h ,wnfi‘ :‘ a. :7 ..‘-.-‘ Y'V' '13-- ".")'
QUJL (Add; DJ 1.4.0. L-Li':'r)t\ OVAu Ami. DIANA."

 

 

 

 

 

 

 

 

 

If we could devise _*~’“'VU._H_ orening strategy such as _____ ,'it
world war Zer swell exempliiy‘“mmfl_fi theme and aims of book, for
our concern ”_“mu_m__u will be with a ~~_ for selecting best sti‘ iC Hog es ,

- ax contexts where the ”~"_“_ 'stratcgy' itself may not

in common use.

The ‘_ of'interest to us _u_ ‘__those in which pe0ple
at crossepurposes: in short, _‘ ..... __“_” situations. The prob3.en\ of to
be gin tlri LS been “““~‘_”_~.recognizab}.y of that type, certainly you and
the are at cross-purposes, as interests are opposed v- in

 

polite way, of course, _' definitely opposed. For we

 

 

 

 

 

 

 

 

 

 

 

 

 

to-coze n you iz.to . - very difficult type of activity, while you
a person with enough troubles , may crave only relaxation
satisfaction of curiosity. "his of interests is essential

_ ,“ptae situations we shall .

Another.eleaent is also _-“_and it 1 present too: Each of
\u;_m;_**_ exert some control over - - ' situation. May ways will

__to you: for one, _~_" _~ may threw the book the cat, thus

11 lngL1PP.~__ _mum the write and the _~. .-_, but at some cost

 

 

property, perhaps some in , and undoubtedly some in _‘

m

 

relations with the ca“. you may skim the _gfparts, and so on.

*mwm

 

are aspects within the -. of the writer, too, as

 

 

 

 

the choice and ___‘”~fi_”“ cf contett -- but it ‘“__ not necessary to

labor __ point. And a further element: appears: Some

sapects . .. Ize si.tua ti on are not the control of either
_fi us; for exam ple, a of events in our and

 

 

extraneous influences during writing and reading periods

YOU ARE EON ASKED TO INJICATE HOW YOU FEEL ABOUT TRE ESSAY YOU HAVE JUST READ.
BETWEER.EACH PAIR OF WORDS YOU WILL FIND EVEN BLANKS. PUT AN X IN THE ONE

THAT REFLECTS YOUR REACTION. HERE IS AN EXAMPLE:

SCIENTIFIC RESEARCH

 

COLOPJ—‘UL x DULL
hammers? ' , x . STUPID
won'rmss x VALUABLE

If you felt that scientific research was very gull, you would put an X in the
first line atithe far right. If you felt thet it was neither lntelllgcntncr
stupid, you‘would put an_X in the middle blank of the second line. And if you
felt that it was only slightly valuable, you wguld put an x in the third line

in the second blank from the right. Now put one XOIn each line of the following

several lines, remembering that you are stating your reactions to the essay you

have read.

HE NATURE 92 STRATEGY

-‘

 

VALUABLE ‘ WORTHLESS

 

 

 

 

 

 

 

WEAK __ __ __ __ ' smonc
gLEAsmr ' __ __ ___' __ UNPL‘EASANT
HONEST . ___. __ __ __ DISHONEST
PASSIVE ___ ‘ _‘ __ __ 15¢;th
UNFAIR “ __ ___ __ ___JrAm
' LARGE _ I. ___ __ __ _. SMALL
CLEAR ’ __ __ ._ __ HAZY
ROUGH ' ' ___ __ ____' __ SMOOTH

 

BAD GOOD

'Do you consider the quality of writing of the essay you have
just read to be good or bad? Indicate you answer by checking

the appropriate box below.
WELL WRITTEN

BADLY WRITTEN

THANK YOU VERY MUCH FOR YOUR COOPE KTION IN THIS EXERCISE. NE
WILL TRY TO GIVE YOU THE RESULTS WITHIN A REASONABLE TIME AND
GIVE YOU OPPORTUNITY TO DISCUSS ANY QUESTIONS OR COMMENTS YOU

UAY HAVE THEN.

Degree Version

 

 

Student's Name

V. A. \

Hess and Major

INSTRUCTIONS

This is a Aspecial exercise which you are required to do during this class
period. The objective is to measure the group's ability to read text of the
kind that‘is in your textbooks and to evaluate your ability to respond to it.
Although you.will not be "graded" on the work you do during this exercise, a
record will'be kept of how'well you do and some judgment will be made of your
ability. It will be prudent, therefore, for you to do as well as you can.

First you will be asked to read an essay of some lZSO’words in length.
Read it carefully but do not try to memorize what you are reading. Reed at
a speed which seems appropriate to you in terms of the text itself and your
own reading ability. Speed of reading is not a matter of interest in this
exercise.

Next you {an be given a new kind of test to see what you have been able
to understand iron‘ the text. When you get to the rest, read its instructions
carefully; they are very brief but important.. .

Third, and _last, you will be asked to answer one question about the text,
your answer being one check mark in an apprOpriate Space. Again do precisely
what the instructions given there say.

Asyou go through the pages of this booklet, 29 2101‘ TURN BACK 39 33 gig-

no m as: ANY QUESTIONS or THE Issrsucross on or yous FELbQIL; s'runsms.
None of then, probably, knows the answer. Read your instructions carefully
and you trill knelt all you. need to tor this exercise.

TURN‘THB PAGE NO" AUO PROCEED. Read the essay and continue on through all

the pages of this booklet. When‘you are finished, hold up your hand.

TIE I‘IATURE OF STRATEGY

It is all too clear at this nauent that there are many, '
ways for e book to begin; and most of those in plain sight
are transparetnly bad . We are tantalised by the thought that
somewhere among them any lie hidden a for having such noble
qualities as these: The readers are informed -- perhaps with-
out suspecting it, though in the clearest prose -- of what the
writer intends to discuss; yet at the same time, it sounds like
the Lorelei calling . Whersupon those readers resolve into two
groups: The first, a large and happy family really, will stick
to the book to the end, even though unimagined adversities in-
pend. Further, this group will alwyas think and speak kindly
of it, and will doubtless have at least one copy in every roan. I
The second group is most briefly described by stating that it
differs from the first; but the book acts imedistely as a espo-
rifio‘ on‘all unpleasant passions, so, as it is sleepily laid
~~ aside, the sole lasting impression is that of a good gift sug-

gestion.

. If we could devise an Opening strategy such as that, it
would wonderfully exemplify the theme and aims of the book,
for, our concern throughout will _b_e_ 31.29. 3 M £9; selectig
133315. strategies, even in contexts where the word -'stra'tey' it-

1"

self may not be in comon use.

The contexts of interest to ours are those in which people
are at cross-purposes: in short, conflict situations. The '

problem of how to begin this book is recognisably of that type,

for certainly you and the writer are at- cross-purposes, as our

,\

' \

interests are opposed -- in a polite way, of course, but
definitely opposed. For we hope to cease you into a very
difficult type of intellectual activity, while you 5: rec-
sonahle person with enough troubles already, may crave only
relaxation or satisfaction of curiosity. This conflict of

interests are essential in the situations we shall study.

Another element is also essential and it is present
here too: Each of us can exert page control over the situa-
tion. Many ways will occur to you: for one, you may throw
the book at the cat, thus irritating both the writer and the
cat, but at some cost in property, perhaps some in self-
respect, and undoubtedly some in deteriorated relations with
the cat. Or you may skim the hard parts, and so on. There
are aspects within the control of the writer, too, such as
the choice and treatment of content -- but it is not' necess-
ary to labor the point. And a further characteristics element
appears: Some aspects of the situation are not within the
control of either ,7°“_,3r_13 for example, a multitude of
events in our pasts and extraneous influences during the writ-
ing and reading periods will play important roles. of course
. this particular problem of beginning the book in a real
optimmn way, has a further characteristic which we shall
henceforth shun, namely, it is too hard e- else we should

have solved it .

The restrictions on the subject matter be so few and
mild, it follows that the set of conflict situations are are

willing to consider is most notable for its catholicity.

There is no objection, in principle, to considering an
I—i~boah contest between Mars and Earth, or a love affair

of the Barrett-Browning type. The contest may be econom-
ic in character, or it may be Musical Chairs. Or it may

be almost any one of the myriad activities which take place
during conventional war. It doesn't follow that we had

a nostrum for strategic ills in all these fields, but there
is a possibility that our offering may as, a method, perform

useful service in any of them.

The method which will be presented is identified by
the catch phrase Game Theory or, time permitting, the m
.93: gm of. Strategy. . If this is your first encounter with
that unlikely sequence of nouns, the sole reaction is prob-
ably: Why? Well, the idea takes their names from the circum-
stance that the study of games is a useful and usable start-
ing point in the study of strategy. That does not really
help, for again we hear: Why? hell, because games contains
meal: of the ingredients common to all conflicts, and they
are relatively mnenable to description and to study. (In-
cidentally, having used the word 'game' to name the theory,
we then call any conflict a game when we are considering it

in the light of the theory.)

. f.
To illustrate the point,‘ let us run our minds over a

Poker game, keeping watch for items which are significant

in, say, a military conflict. You and four others are thus .

studying hmnan nature, under a system of rewards, you hope.

we note at once that the players have opposing inter-
ests; eachwwants to win and, because the'winnings of

one are necessarily the losses of another, their inter-
ests are oppose. This provides the basis of conflict.
We observe that some elements of the action, being per-
sonal choices, are completely'within your control. And
the same being true for each player, there are elements
which are not within your control; worse, they are con-
trolled by*minds having interests inimical to yours.
Finally, there are eleuents of the game that are not,
under the rules, within the contrOl of any player, such
. as the order of the card in the deck. These elements
may he thought of as being controlled by Nature -- who has
a.nassively stable personality, a sonewhat puckish atti-
tude toward your important affairs, but she hears you no
conscious malice. These are all surely familiar aspects

of any conflict situation.

Another characteristic is that the state of informa-
tion -- intelligence, in the military sense ~- is a factor,
and, as usual, is an imperfect and hence troublesome factor;
we don't know‘whst the other felloe's hole card is. There
is also the bluff by'which you, or the opposition, give false

evidence regarding intentions or strength of forces./ Other

similarities will occur to you; people even get killed,

0

occasionally.

But the analogy should not be pushed too far. You can

think of many aspects of z-ra>:;-’Eare which are not reflected

in Poker. One tank will sometimes kill two tanks, in a
showdown; whereas a pair of Jacks always wins over an Ace-
high hand in the showdown. 0f coarse Poker could be modi-
fied to make it contain showdown possibilities of this kind,
say by riding that an Ace is superior to any pair, up to
Jacks, whenever anybody's wife phones during the play of a
hand. But the fact is that the game don't exhibit all the
complexities of warfare and of other real-life conflict
situations -- which is precisely why they are usable start-
ing points for a study of strategy.‘ In the early stages

of developing a theory it just is not possible simultaneously

to handle very many interacting factors.

It is probably clear, then, that games do contain some
of the basic elements that are present in almost any inter-
esting conflict situation. Does it £01m; that we can learn
useful. things by beginning a study with them? Not necessard
ily. It may'be that _military, economic, and social situations
are just basically too complicated to be approached through
game concepts; This possibility gains credence from the fact
that the body of Game doctrine now in existence is not even
able to cope with full-blown real games; rather, we are re-
stricted at present to very simple real games, and to watered-

- cl

dot-m versions of complicated ones, such as Poker.

WITH FEM OR PH

.7 vv vv: M7.) fi'fl: r'
:Cilag I.Q.-u Ln {114.311. .3

YOU COHSIDER HOST APPROP“IRTC. TRY HOT TO LEA

QUICKLU; DO NOT LhEOR OVER RWY BLANK.

 

Ii; 111$:

78 AMY OUT.

FEXT BELOW HITii TEES hOP \D

10FKI.LOHG FAI”1.LY

If we could devise a; “ Opening strategy szch so wvwfiédfi,n,. it
would wonderfully exemplify theme and aims of a book.
“scan. M2 m.vz.~...\..‘a-~ fix“:
for our concern "d'will be with a _ ,, for "v"~’1“p 1
strategies, in contexts where the 'stratcrv‘ itself
nun.amrma ’ ’
may ‘A,___A‘h__be in common use.
contexts of interest to a Hau‘ar~ Chli: h>.l-\m
are at cross-purposes: in ., conflict sjl.—t.nnx. The
“W
problem how to begin this .. is recogniz.1ul' of t1-e:
53W
‘ . for certail .ly you (And .41.” te‘ 1? are 65.1.. CL“.‘.~3 "1)\‘LA- t1k/Lfit". ~') ’

‘ our interests are Opposed -~

 

but definitely opposed. For an»

difficult type

‘.-“A—‘A“-

.able person with enough

satisfaction of curiosity.

-.= .agv"

in the situations we study.

Another element is

MEL-Cal) Jan-Sh“

can exert some control

‘L‘--‘--‘A‘

Each of

ways will to you: for one,

the cat, thus irritating

WW

some cost

preperty, perhaps some in

some in relations with the
M

«Ln-cum

hard parts, and so " A_.

--‘A A -

of the writer, “

but is not necessary to,

W“ “M3635

characteristics element appears: Some

within the control of “1 you or 1;

15.3.1". a 4.’ a";

a ..J._,- . ._. .'. . ‘ . .‘pl 4 ..
e‘J-‘JLZLu (31.1.0 Pan-SL3 curl ‘15. 56.1.. . 4'. ;.

«and. 33“” m

Mung-5.1.1.3.:

hose to cozen you
a . e
intellectual activity, while you
already, may crave only

'conflict of interests are

our ontial and it is

the writer and the

. a‘fl‘rzag“ C. ‘\ Ila'fi’

WW

\
There are sapsots within
, such as the choice
a.
the point.

11.. .‘...L:‘q~'.'_".‘ . 3.13.1

for
.4..-

a polite way, of s

“.11).:- 42m}

a very

:m’. 1m

1‘? \l (1‘90“-

OP

“WW ' " '

M: 2.2:.» -~ an

here'too:

MILL—:2 3.31 7H.)

the situation. Huey

ma 3! throw t1..’2 book.
. .. ‘n.'i cl“:¥nln‘:

, but at

and undoubtedly

"3":

. Or you :9 eye in

control

t N.tnont of content --

m 2.3: 11.11:.“

And a

«A n.7h.x:uasa. L.”

of the sitar t1 ovn are

union’DoL—h' Q‘t—w. 3

H, :1 :nujfirl'tude of

YOU ARE E03 ASKED TO ifiulflflffi r53 EOU PEEL ABOUT TAB ESSAY YOU HAVE JUST READ.
BETWEEN EACH PAiR OF WCRDS YCU WILL FIND SEVEN BLANKS. PUT AN X IN THE ONE

THAT REFLECTS YOUR REACTION. HERE IS AN EXAMPLE:

 

 

 

 

 

 

 

s_m-rrxnc stsAagg
COLOIU-‘UL ___ __ ‘ x DULL
INTELLIGENT ___ '______ - x STUPID
wonnmss x VALUABLE

 

 

 

If'you felt that scientific research was very dull, you would put an X in the
first line at the for right. if you felt that it was neither intelligentror
stupid, you would put an X in the middle'blank of the second line. And if you
felt that it was only slightly valuable, you would put an X in the third line

in the second blank from the right. Now put one X.in each line of the following
several lines, remembering that you are stating your reactions to the essay you

have read.

us NATURE 93 STRATEGY

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VALUABLE ___ __ __ ' wonrmzss
WEAK __ ___ ___ ‘ STRONG
pumsmr ~____ __ __ ' UNPLiEASANl‘
HONEST ___'__ __ ___, ”1 ' I ntsnomzsr
PASSIVE ___ ___ ___ , . Active
UNFAIR __ ___ __ ‘ FAIR
LARGE ___ ‘___ __ 2mm.
CLEAR ___, __ __ _ . 1»sz
ROUGH __ __ __ ‘ suoom

 

 

BAD GOOD

' Do you consider the Quality of writing of the essay you have
just read to be good or bad? Indicate you answer by checking

the appropriate box below.
WELL WRITTEN

BADLY WRITTEN

THANK YOU VERY MUCH FOR YOUR COOPERATION IN THIS EXERCISE. WE
WILL TRY TO GIVE YOU THE RESULTS WITHIN A REASONABLE TIME AND
GIVE YOU OPPORTUNITY TO DISCUSS ANY QUESTIONS OR COMMENTS YOU

MAY HAVE THEN.

 

19.9:

AAA... _--

1
I
l
l
1
1.
i
4
1
.
5
c
s
3

 

 

IfilYi ‘

W
l‘ WRITE IN THESE BLANKS FILL .H‘I THESE BLAUKS Version 2

:ee Version ”A ._
Student's Name

 

 

Class and Major

INSTRUCTIONS

This is a special exercise which you are required to do during this cross
:riod. The objective is to measure the group's ability to read text of the
nd that is. in your textbooks and to evaluate your ability to respond to it.
though you will not be "graded" on the work you do during this exercise, a.
:ord willo'be kept of how well you do and some judgment will be made of your
.lity. ‘It will be prudent, therefore, for you to do as well as you can.
First you will be asked to read an essay of some lzsolwords in length.

d it earefully but do not try to memorize what you are reading. Read at
peed which seems appropriate to. you in terms of the text itself and your
reading ability. Speed of reading is not a matter of interest in this
'cise.

rext you (0111 be given a new kind of test to see what you have been able
nderstand from‘ the text. When you get to the rest, read its instructions
fully; they are very brief but important... .
11rd,, and‘lest, you will be- askedto answerone question about the text,

answer being one check mark in an appropriate Space. Again do precisely

the instructions given there say.

 

you- go through the pages of this booklet, 25; NOT TURN BACK :9 ANY PRE-

3 mos.

/

nor as: ANY QUESTIONS or THE msrsucrons OR or mun EELLou ssussnrs .

/
o—d,

pf then, probably, knows the answer. Read your instructions carefully
.u still know all you need to for this exercise;

N THE PAGE M AND PROCEED. Read the essay and continue on through all

gee of chi. booklet. Uhenxyou are finished, hold up your hand.

 

'1“- w «Hr-m

A o f "F
L.£i.*.d.sE U3 STRLRTHGY

It is all too clear at this moment that there are many
ways for a hook to begin; and most of those in plain eight

are transparently oau. fie are tantalized by the thought that

somewhere among them mas lie hidden a few having such noble

Odie -J '.

qualities as these: The readers are informed -- perhaps Witl~
out suspecting it, though in the clearest prose -- of what the
writer intends to discus ; yet at the same time, it sounds like
the Lorelei calling. Whereupon these readers resolve into two
groups: The first, a large and happy family really,‘will stick
to the book to the end, even though unimagined adversities im-
pead. \Further, this group will always think and speak kindly
of it, and will doubtless have at least one copy in euery room.
The second group is most briefly described by stating that it
differs from the first; but the hook acts immediately as a sopo-
rifle on all unpleasant passions, so, as it is sleepily laid
aside, the sole lasting impression is that of a good gift sug-

. gestion.

If we could devise an opening strategy such as that, they
would'uonderfully exemplify the theme and aims of the book,

for on}: ggzzcern throughout will he with a method fog: selecting

 

 

w“

best strategies, even in contexts where the word 'strategy' it-
self shall not he in common use.
The contexts of interest to ours are those in‘which peeple

late at cross-purposes: in short, conflict situations. The

roblen of how to begin these pages are recognisably Of the type,

"3

for certainly you and the writer are at cross-purposes, as our

     

r '"..

(tiv'
-.g

' ' ‘xv""""ashoo
Codi->1 II n I...
. .

 

   

.4; -

1"! “ o‘.
>0 . U‘
0
M19 yoogw.‘

‘I

I

““o‘

    

w ""1"
19-4., .-

<Ie~
be. ' ‘
..-AA-l‘

inter etc are ogposed ~- in a polite way, of course, but
finitely opsosed. For our hope to cozen you into a very
difficult type of intellectual acti‘ity, while you a rea-
sonable person with enough troubles already, may crave only
relax tion or satisfaction of curiosity. This conflict of 9

interests are essential in the situations we shall study.

Another element is also essential and it is present i
here too: Each of us can exert Eggg'centrol over the situa-
tion. Many ways will occur to you: for one, you.msy throw E 4
the book at the cat, thus irritating both the writer and the
' cat, out at some cost in property, perhaps some in self- w
respect, and undoubtedly some in deteriorated relations with

the cat. Or you may skin the hard parts, and so on. There

 

cvvfiuruv-

'_ V‘Yv
O'd‘.p'flnl“‘ O'C'Atot-
.44 .4 A ,vo‘.

I ' A

1

are aspects witsin the control of the writer, too, such as L

4

the choice and treatment of content -- but it is not necess-
ary to later the point. And a further characteristics element - H
- appea‘s: Some aspects of the situation is not within the 1
.control of etther of you sad I; for example, a multitude of i
events in our pests and extraneous influences during the'writ-
ing and reading periods played important roles. Of course y
ethis particular problem, of beginning the book in a real
optimum say, has a further characteristic which we shall
. /

heaceforth shun, namely, it is too hard -- else‘we should

have solved it.

The restrictions on the subject matter be so few and

zuild, it follows that the set of conflict situations‘we are

‘0

selling to consider are most notable for its catholicity.

.ole, to consider an h-bomh

Barrett~3rowring type. The content may be economic in chat-
acte“, or it may be Husicel Chaits. Or it may be almost any
one of the myriad activities which take place during conven-

"allow that we had a nostrme for

.4
‘4
H
8'.
o
I"!
fa
9
0
0
V)
I}
3‘
3')

tional
strategic ills in all these fields, but that 13 a possibility
that ours offering may as e.method, perform useful service in

any of_them.

he method which will he )resented is identified b*
1

fi'

he catch phrase Gene gheoqy or, time permitting, the Epcot:
5 Gates f’gggetggx, If thia is your first encounter with

9

that unlehely sequence of nouns, the sole reaction is prob-

f.)

bly: why? well, the idea take their names from the circum»

stance that the study of games is a usefully and usable start-

p.
11:

C»?

’0
0
we
:5
f?
‘5'.

n the study of strategy. That does not really

4.
V

4" ~

A:

Lclp,fo; again we hear: Why? Well, because games contains

-‘ 3 3: lo 7~ .9 a. -A .° 9 .‘o «a? -_ n o . a gag .. ‘.

many 0; the,tnetedteats coemon to all confllcts, ena they are
relatively amenable to description and to study. (In~

cidentally, having sed the word 'gane' to name the theory,

“the. n‘ l l .E'} F‘?M‘—n“ “A.

we then call any conflict a game when‘we are consiflering it

in the light of the theory.)

To illustt te the point; let us run our minds over 3
Poker game, keeping watch for items which are significant
in, say, a military conflict. You and four others ate thus

studying h'san nature, under a system of rewards, you hope.

 

We note at once that the players have pposirg inter-

ests; each wants to win and, because the winnings of

ests are o.pose. Tnis province the basis of conflict.

We observe that some elements of the action, being per-

sonal choices, are completely within your control. An
the same being true for each player, there are elements
which are not within your control; worse, they are con-
1

trol ed by minds having interests inlmical to yours.

‘Finally, there are elements of the game that are not,
under the rules, within the control of any player, such
as the order of the cards in the deck. These elements
may be thought of as being controlled by Nature -- when,
we believe, a.massively stable personality, a somewhat
puckish attitude toward your important affairs, but who
bear you no conscious malice. These are all surely

familiar sapects of any conflict situation.

‘Another-characteristic is that the state of informa-
tion -- intelligence, in the military sense -- is a factor,
and, as usual, is an imperfect and hence troublesome factor;
we don't know what the other fellow's hole card is. There
is also the bluff by which you, or the opposition, give false
eVidence regarding intentions or strength of forces.f Other
similarities will occur to you; people even get killed,

'occasionally.

,o

But the analogy should not he pushed too far. You can

 

«Ca
(GO-t.

slewimm' JLWnMH‘H-M

' .V '. ..., .,'§:..;.¢. Add
on- .4-o¢— A - - -
coura- J'MZN~

rm

'u', 1,...

all .

 

IQ —'n
al.35-

  

‘l'... -
., . _

  
  

JP” :1

'vr_--q.. -.-

.'.‘.7V- I? a . . . -- h ., ...,r; -. J ' ‘ . I. .- .7 . 1
Ldlflm oi naiy asvects o! matinee union are not railecten

in Poker. Ore tank will sometimes kill two tanks, in a
shout wn; whereas a pair of Jacks always wins over an Ace-
high hand in the showdown. Of course Poker could be modi-
fied to make it contain showdown possibilities of this kind,
say by ruling that an Ace is superior to any pair, up to
Jacks, whenever anyhody's wire phones during the play of a
hand. But the fact is that the game don't exhibit all the
complexities of warfare and of other real-life conflict
situations -- which is precisely why they are usable start-
ing points for a study of strategy. In the early stages

of developing a theory it just is not possible simultaneously

to handle very many interacting factors.

It is probably clear, then, that genes do contain some

Hi

0 the basic elements that are present in almost any inter-

’-

esting cont

lict Situation. Does it follow that we can learn
useful thirgs by beginzing a study with them? Not necessar-
ily. _ t may'be that any military, economic, and social
situation are just basically too complicated to be approached
through game concepts. This possibility gains credence from
the fact that the body of Game doctrine new in existence is
not even able to cope with full-blown real games; rather, we
are restricted at present to very simple real games, and to

watered-down versions of complicated ones, such as Poker.

 

 

 

 

V'Vfi‘ll fi'wp br‘ ”"'T" «o- v.~- . v - ,.. ‘ , . . . . . . ,
. _ ,A , , ‘ y! _ . , . . [I . x V \ I
"J. LU L ul- \)n\ A .’..u(‘. .‘J I I '7 _" ‘. ‘ . n ', 2, ' 1 . i; — . . .
'7’." ,.‘ .Iyp..‘.‘...‘ , , ,‘,‘ . .'. ‘ A .1...._.- , ‘1. ‘,. b ,- < ‘ . p a
.‘ U \4K'hv3nl-«I‘x 1’,\)._ . . ..'- -.".. . I... ..' x '1 ‘ .'.~- . 1 .. a. . ‘
£)v ("1" "' . ‘l ’ ",".|; f I" r‘ y. 5-"? :_ \ v ~ I y - .1;
u - up...“ , -wj 1.\.' UK},-‘~}1\ u ‘. “A . s ' .. ‘ ~ ;
iv‘ ’ .J . U .
A ‘ ‘ ,5 f' \a'.-' ' l . . H ,g " ‘0. . -.' V .I . -
~-\' - nay-J. " '
J"‘fx\;° 'V'0‘7r‘ "a"! ”.Wl . I - . v- -' . ->- ‘ ‘ ...I- . '
l- a H‘J .. \1 -J4_ ‘ .‘...._: . . -. L“ . .. , .., ( \.-. v an 4 ,
a. ‘- 'a _ 'A‘ - n
I ‘ ‘ I
1“, l‘ “ -, ‘~:‘ . ‘ a ' .n v -. .
L-.'. U' \(..‘C,I:‘U w _--1 4‘ .u -‘ .' .
‘3.»- -;: .-.l‘l.‘ -- — .
. ~ _ ‘,' . , p ' .i- :r - '
.-... ‘.»‘ .o . . ‘ .' -
“kn-c. .9..." ‘.:‘_:‘.._'- \ . . H a; .-
.
, -_ . $ .‘ ‘\ ‘50 1‘ ‘ ,~‘ ' .
~|u"J../. u w. .4.” g'._‘...-( _. .- r-..
-\"-_ . 'J-J ':~ '..'_‘ Edi...
'fl‘" ,- . . ' ‘ ! ° ' '
‘ . - ,w- . n ‘ , \ . " ‘\
k .D’ All! .0 I'. ‘- I \y' ' (.a ‘Jx\_ 2r ‘. ‘ r‘ .V...
.._ .. . . I ‘ » J
. _ _
u
L \' .r -I ‘ '3 .~- '| ‘a-r v'zr . - -‘ ‘ . us- .3. "0‘ ' ' . ‘.
_ x.“v Con.‘..r . "“ i":} \0.‘ ¢ _' § 3. ‘J 1111:} - - 1‘ b j- - ' .
In". 1...; 0.6-431.2.-3. it; A." . ._ § , . . . V .3
\r‘ .(. _ .0 . , g‘ _ 1 ‘ v n o ‘. ,-
‘ ' .4- .. yr - ~ - ‘ I , a h. ..
.. I. . ...‘J 14-' ‘ '1‘ .iss 'J\; 1. ' d .. .J' . ,'- s .I .' ‘ ‘ l ‘.
..-_h:,‘_. .-rl‘. :-; .6 c. '..._..‘.\‘a.-; >
"“"Vv"': 3."! urn \ ‘ ‘ .. .
‘ ‘. . . . . . . _ ‘
. (.A—L as..- NJ.) J x). ’-‘ .u u) . u. . . u- . J~.'.
_ ,,,____...\..:.- 53.3.; .4433- :_, ‘ . . L ." ‘ ‘
I . '-
, ...-.'..-p.. ,‘Q ,.. ..-“.... ‘w . ". . .‘. ._ ., . ,
.A.L".-..-,’1u£.n c.1533 Oi-lll‘E-L'J'J. "" 1:1 L- 'u. . o, ('3 L- - 4 , .
\ s.-. .~.:_.' .'..-'._ 3."- '.... ‘ ‘ ‘
" ':' .llv. ' .- 45f. " 4‘ 'o l u w I '
,_ . . . _ ,. I . .. . . ,_ , . . . .. ~.
13A.LII.LLU.L> (Jet’s/“CU. 10L (.x K, - -; ‘,' .11.. ..l ..w‘
‘ ‘” -‘_.‘.."."~.:;..'.—T‘A.‘.. - - - I V
I . $— A ‘ . .‘ f— ‘ I ‘I ’ q 8 ‘
.. - .g --‘- '- "‘-‘- '1 “v. V‘ -'-1 ..-'-n .n < w .l -' .- o-
...,.1 .5 L’g’d-Lx .— ' "\I (J). ‘l<a '. J.» oLLJ s ‘w !.4._..‘. 3’ \. h. L
. -¢L_"—'A:‘--'U .'.\.':g.'l..v._3 r. -' . -" A _ 7.-.~_ 2
‘ ‘ ~ . .
, '."_I‘ . J~v..\u‘P\1 I" Q - 1' ~ - ~02 n . ~ ,' , \V- ‘
‘.A‘,Ak."..‘ ‘\;L.Ll.~\.s.: l; v- ’.' (‘nu II .. ‘- ~A ~“ 1" .'-x .\"'- “ no
" A“_:'_J.'""0C~£§.:-” —'L."-° . . ugh _-- '_ _. -
f- ‘0 v '\u C ‘ ‘ 'I ‘b \ , q ' ‘
- ~~,-‘ (V ' I ' of - ‘ .' ‘ " ‘ uv'fi - N ‘ ‘
L CL‘L..0nJlCE’. ALIL») ( A. *l.gv¢£,. .-' g} \ ','.\-c'." ,‘L ,_
¢:x.zq:‘.;.;- Axxrtchf. .:. -
g .- u. o -'- -: on... f< '-,—., 'u’r‘ u- ’ 1
..l \l:n o-\'l.‘-) ‘O\.a U’JQ‘.L-- o
.u, n > I’ ‘t ‘ -‘ l ‘
“DA-:- V. . -..‘h unfi—v .-
5 1 '. ‘ A
. . -, - . ." g 3‘ .r ‘ : -‘ , \ 0,-7. ,: a . .n .. . , . . K .
l..|k)t_£l ‘3. ‘a-me. ...1 \.:3.«..‘... ;,u\.'\ ,gg, .L.. . , H; g
. - . .. .‘ k . \ Q -‘~ . . . o L
n-.- _. - -._¢..-. . ---, ...- -.. .._._.~-. 5
s I. I -‘ ' I
, ~‘ — -- n1"? — '«1- ' , G I .m'r . ' v- - y » “. . .
.-‘.al.‘ tJ-L “D‘- . buod‘ b u «cud (wk-6b.; C". .‘~ '. '- '-. t. J.
u- :-'.'.-:'.-'l Lh.:.;-i ...1...’ .— ." _ 1:3. ‘7 .'\ :— 4 ‘.
~ . ~ ’ a 1
— ‘ .' . ‘ g-, u . . . - ~.- , 9.. . . . a . ,‘
. u, 1". J. ‘ " “~‘.. ..a.’;‘ ‘t;|.- 3 0‘5' . \Ll‘flfis. s
‘ ...L..- .2.’;‘. ‘2.“9435‘ -’ J .. _ § ‘-'..-...'. -.‘_\”-_ ,,« . -
. , \ V
.' ‘ . «.1. _' ' - . .. ...) . . ’ -. ,f.'
‘d \z. “ .‘. ‘ I. , - ..IA\~ |-'4 A;
‘ \.-'\ . c . .. y
‘5) (”cl ..‘.. .~-.'-'\ ,I,.u:v'l r» r . '. o a : T. -y. N z ‘
P.c.\. .. 1.8.. L‘.-Ld\.(. t.) ‘2 :‘..;.a,.,>’1_, _\ , —-l \ k‘n
u;.‘41.1‘d.“;‘1‘»'3-a¢$4 L.‘ . q, ,- _." _ '..' .n ‘ 2'
v a n ' . . \ I 1 .
1‘ .,. «w, ....,.-‘ “"’.‘ ,-'-.- a.."¢- ,... . .,', .
-_ 1‘, 4.9 1.33.31: L -‘-(“At.l., -‘4 . ‘. \.'. L1; .. git l L g ..P‘.‘ 1.
-¢;'.‘...'.;a.2-";:’-A.:u13 ..'-."--‘..’,-_’L 2‘: ...J;-. '
I
r}- i "i " r0 5‘ Q a ' I . '«‘ ~ ' ‘ I‘
_,. . , .7 ""1," a.- ,_ .. . ... ,._ _,' _
1. Lb 9 513“ \a‘- (Jo-o 's‘.-- L. f» ..'..\.-n ~..~.~,)1.‘.ls Lbl\,‘.
u’.:..$."._".."..'.‘..“; ;__.', \__ _. _;. 5. 1. .. V-
. H .. J.“ 1. .‘c? «5 ~ I --- I n - ‘ l n 'o ‘ .
. _ ~ _. '. 4, ‘ . , .'. .. . . .
‘ 9 7‘ .\ - “ ‘ L".(‘ r “"" V I. " "(“ _ . ‘ ‘/V" ‘r‘- l" ‘— ... t. umx :- \.
“L“..a—Z-.:.~ - _.. .Jc. -~ .7- . .. '_,‘ :L . ..-
Q ‘ O | n . .- .
- . r -\f'( . ~ (A, ...n f . ,I‘-.~, "‘ , . ,' - . ‘
- -.)v--‘C"d ‘~‘0\?:~t‘ 1: L‘.) -.-‘. ~"..".‘ n—"- — t .u a I ' ' » ‘ » L.
’ s H'- 7 - -" ‘ . .
D
s . n a
.h, o. ‘0 . ( ' ‘pv‘ - ~ ~.\ " '« I I: - ~ _ x u
, I . . -- 5: O ‘1 ’ .- ‘ § \ .4 . a . , 4 \ . , ,. _.
. .'.j- . , . ‘ . - -
. a 0. 3; . l
- _ _‘ a .
go . ...' . ' b ' ' \ \
' . -. U x: . : .

 

 

 

. \ ..
_ J)
.
.
-A‘ o ‘
. I
. .. a
'3 .
1‘ . '
.‘ . c
\
d
.'.-J.‘7-- ...'._
..
\ ‘0
..
_.
.v
3
LL.
8
”r- ‘
.‘ . . (J51
O
‘ ..
‘J'I:
I‘r
,1 1; “ ‘..'x:1
;o
L..-
.
.m . \‘7

 

 

   

I‘tl'
‘1"

   

   

.av.“
‘54 1‘.
. .
‘J‘...~, '

    
   
 

-
.“

  

I

4

  

f-
*___'O l -_.A

4

   

9
‘1'

   

    

‘ .

   

II“, If

c.44‘

MAJ}

   

0".
.C'.

..I,

    

    

   

   

'. . L
4.0:”.

   

t

  

. ‘
O

      

“M 3..

    
 

-

‘4'

    

"m
J‘-

  

H

 
  

.53

  

.4 h
’3

q
n
-

   

2‘2}

 
       

‘-
b

“vr‘—I
‘

    

o

     

-
‘

   

        
 

'

  

1!

'OU ARE NOE-1 ASKED TO INJICATE HOW YOU FEEL ABOUT THE ESSAY YOU HAVE JUST READ.
E'ITIEEIJ EACH PAIR OF WORDS YOU WILL FIND SEVEN BLANKS. PDT AN X IN THE ONE

IIAT REFLECTS YOUR REACTION. HERE IS AN EXAIH’LE:

 

ggIENTIFIC RESEARCH 5

COLORFUL x DULL
IllTELIJGEIJT ' - x . STUPID ;
won'zmass x VALUABLE '

f you felt that scientific research was very dull, you would put an X ln the
.rst line at the fer right. if you felt that it was neither intelligentrcr
upid, you would put an X in the middle blank of the second line. And if you

It that it was only slightly valuable, you would put an X in the third line

4”
av ' -

0 O .-
rat:

the second blank from the right. Now put one X.in each line of the following

4.. ..., .

.0 "
.6“.I~
W

 

 

verel lines, remembering that you are eteting your reactions to the essay you kgg
ve read. fifi
.fi:
_ z!
.- Egg UflTURi‘_§ STRATEGY j '8
VALUABLE __ __ __ __ __ ’_ __ WORTHLESS 3
WEAK __ __ __ __ __ __ __ ' STROHG ll:
IjozAstT __ __ ___ __ __ __ __ UNPLEASANI‘ :j
HONEST __ __ _____ ___ __ __ __ DISHONEST I:
PASSIVE __ __ __ ___\_ ___. ___ __ 4941?}: ll
UNFAIR __ __ __ __ __ __ __ "1mm :
‘ LARGE, __ _-___ __ __ __ __ __ SMALL . A;
CLEAR. .__;fl ____ ___ .___ .___ .____ ___. HAZY
ROUGH __ _ __‘____ ___ __ __ __ z-zooru
BAD “___ __ __ __ coon

 

 

' Do you consider the quality of writing of the essay you have

jusr mud to be good or bad? Indicate you answer by checking

the appropriate box below.
GSLL WRITTEN

BADLY WRITTEN

TUKHK YOU VERY MUCH FOR YOUR COOPERATION IN THIS EXERCISE. WE
WILL TRY TO GIVE YOU THE RESULTS WITHIN A REASONABLE TIME AND

GIVE YOU OPPORTUNITY TO DISCUSS ANY QUESTIONS OR COMMENTS YOU

MAY HAVE THCNo

 

 

 

 

'RITE IN'TKESE BLANKS FILL IN TIESE BLANK Version 3

Version‘y
Student's Name

 

Class and Major——

INSTRUCTIOES

This is a .special exercise which you are required to do during this class
iod. The objective is to- measure the group's ability to read text of the
:1 that is. in your textbooks and to evaluate your ability to respond to it.
trough you will not be "graded" on the work you do during this exercise, a.
er*will'be kept of how'well you do and some judgment will be made of your
Lity. It will be prudent. therefore, for you to do as well as you can.
?irst you'will be asked to read an essay of some 1250)word3 in length.

I it carefully but do not try to memorize what you are reading. Read at
reed which seems appropriate to. you in terms of the text itself and your
reading ability. Speed of reading is not a matter of interest in this
’cise.

'0

text you will be given a new kind of test to see what you have been able

mderstand from‘ the text. When you get to the rest, read its instructions

fully; they are very brief but important..
'hird, andmlast, you' will be asked to answerone question about the text,
‘ answer being one check mark in an oppmpriate space. Again do precisely

the instructions given there say.

 

8 you. go through the pages of this booklet, 92 NOT TURN BACK go .3113; 331;-
0 nor ASK ANY QUESTIONS or THE msrnucrons on or your: E‘ELlei; swarms .
of then, probably, knows the answer. Read your instruct/ions carefully
you will know all you: need to tor this exercise.

URN ms nos saw AND mosses. Read the essay and continue on through all

pages of this booklet. When‘you are finished, hold up your hand.

0s -..

n to...

1...».-4-
.

fif-

I Q-If.
t'r-M-M'

‘ --— "V“ ‘18-: 1.
#o.-'.J ,, , ‘ ‘ - ' - o
. ' -.-.I.. _' - "‘ '
G
.' v ,

 

H’s?
" .‘o'.

o 1“

  

on... [an 0’!"
‘~-« -4 g '

rib
.1‘
'3
J:

t
I
5.:
"I:

vr ‘7 ‘ ‘01“ (""73 \"’”~"‘17
TLLE J..l“:.TJ;tih OF I) Lt‘iim'ul.

It is all too clear at this m ment that these are many
ways rot a hook to testes and most 5 those in plain sight

are transoarcntly had. We are tantalized by the thought that
somemhcre among then gay lie hidden a few having such noble
quali“ies as tnesezv The readers are informed -- perhaps with-
out susgecting it, though in the clearest prose -- of what the
writer intends to discuss; yet at the same time, it sounds like
,thc Lorelei calling. Uhereupon those readers resolve into two
groups: The first, a large and happy family really, will stick
to the book to the end, even though unimagined adversities im-

pcnd. Further, this group will always think and Speak kindly

of it, and will doubtless have at least one copy in every room.

 

“A.

The second group is most briefly d scribed by stating that it ;3

3iii-fees from the fi~stg but the book acts immediately as a sopo- g :2:

rifle on all unpleasant passions, so, as it is sleepily laid :, ii

asite, the sole lasting impression is that of a good gift sug- J‘ :5

gestion. W g
I

If we could devise an opeing strategy such as that, they ' y

'I

' “econ-Mm: A.
A— p o
4..

”I fr.

would wonderfully exemplify the theme and aims of the book, ,

-.v- M mam ~‘DIIM ‘OUvm ...h“ ”In...“

for on? concern th393§§9“t will he;with a method for selecting

host strategies, even in contexts where the word 'stratcgy' it-
/,I

'self shall not be in common use. -~

The contents of interest to ours are those in which people '
1

are at cross-purposes: in short, conflict situations. Tee

C

problem of how to begin tnese pages are recognizable of that type,

for certainly you and the writer ate at crossnpurposes, as our f

interests are opposed -- in a polite way, of course, but
definitely opposed. For our hope to cozen you into a very
difficult type of intellectual activity, while you 'a rea-
sonable person with enough troubles already, may crave only
relaxation or satisfaction of curiosity. This conflict of

interests are essential in the situations we shall study.

Another element is also essential and it is present
here too: Each of us exert 33393 control over the situation.
Many ways will occur to you: for one, you may throw the
book at the cat, thus imitating both the writer and the
\x‘cat,Kbut at some cost in preperty, perhaps some in self-
respect, and undoubtedly some in deteriorated relations with
the cat. Or you may skim the hard parts, and so on. There
are asPects within the control of the writer, too, such as
the choice and treatment of content -- but it is not necess-
ary to labor the point. And a further characteristics element
appears: Some aspects of this particular situation is not I
within the control of either of you and I; for example, a
multitude of events in our paste and extraneous influences
during the writing and reading periods played important roles.
Of course these particular problem, of beginning the book in a
real Optimum‘way, have a further characteristic whichuwe shall
henceforth shun, namely, it is too hard -- else we should have

. 0"

solved it .

The restrictions on the subject matter be so few and mild,
it follows that the set of conflict situations we are willing

to consider is most notable for its catholicity.

 

 

There is no objection, in principle, to consider an H-bonzb
contest between liars and Earth, or a love affair of the
Barrett-Browning type. 'i‘he contest may be economic in
character, or it may be Musical Chairs. Or it may be almost
any one of the myriad activities which takes place during
conventional war. It doesn't follow that we had a nostrum
for strategic ills in all these fields, but there is a
possibility that ours offering may as a method, perform use-

ful service in any of then.

The method which will be presented is identified by

the catch phrase Game Theory or, time permitting, the m
3; gm of Strategy. If this were your first encounter with
that unlikely sequence of nouns, the sole reaction is prob-
ably: Why? Well, the idea take their nanies from the circum~
stance that the study of games is a usefully and usable start-
ing point in the study of strategy. .That does not really

" help, for again we hear: Why? Well, because games contains
-many of the ingredients common to all conflicts, and them
are relatively amenable to description and to study. (In-
cidentally, having used the word 'game" to name the theory,
‘we then call any conflict a game when we are considering them

in the light of the theory.)

\ _
To illustrate the point, let us run our minds over a
Poker 3. ..e, keeping watch for items which are significant

in, say, a military conflict. You and four others are thus

studying human nature, ‘under a system of rewards, your hope.

We note at once that the players have opposing inter-
ests; each wants to win and, because the‘winnings of

one are necessarily the losses of another, their inter-
ests are oppose. This provides the bases of conflict.

We observe that some elements of the action, being per-
sonal choices, are completely within your control. And
the same being true for each player, there are elements
which are not within your control; worse, they are con-
trolled hy minds having interests inimical to yours.
Finally, there are elements of the game that are not,
under the rules, within the control of any player, such
as the order of the cards in the deck. 'These elements
may be thought of as being controlled by Nature -- whom,
we believe, has a.massively stable personality, a somewhat
puckish attitude toward your important affairs, but who hear
you no conscious malice. These are all surely familiar

aspects of any conflict situation.

a.

Another characteristic is that the state of informa-
tion -- intelligence, in the military sense -- is a factor,
and, as usual, is an imperfect and hence troublesome factor;
‘We don't know what the other fellow‘s hole card was. There

is also the bluff by which you, or the opposition, gives false
evidence regarding intentions‘or strength of forces.//Other
similarities will occur to you; people even get killed,

'occasionally.-

’I

But the analogy should not be pushed too far. You can

 

 

think of many aspects of warfare which are not reflected

in Poker. One tank will sometimes kill two tasks, in a
showdown; whereas a pair of Jacks always wins over an Ace-
high hand in the showdown. Of course Poker could be modi-
fied to make it contain showdown possibilities of this kind,
say be ruling that an Ace is superior to any pair, up to
Jacks, whenever anybody's wife phones during the play of a
hand. But the fact is that the game don't exhibit all the
complexities of warfare and of other real~life conflict
'situations -- which is precisely Why they are usable start-
ing points for a study of strategy. In the early stages

of developing a theory it just is not possible simultaneously

to handle very many interacting factors.

It is probably clear, then, that games do contain some
of the basic elements that are present in almost any inter-
esting conflict situation. Does it follow that we can learn
useful things by beginning a study with them? Not necessar-
ily.. It may'be that any mdl_tary, economic, and social
situation are just basically too complicated to be approached
through game concepts. This possibility gains credence from
the fact that the body of Game doctrine now in existence is~
not even able to cope with full-blown real games; rather, we
are restrict at present to very simple real games, ena to

watcred-down.versions of complicated ones, such as Poker.

V{IFH PEN OR PENCIL, FILL IN THE BL.HKS IN THE TEfiT BELOW ”IT” TEE WORD
ENNJ CONSIDER HOST fiPPYOPRIATH. TRY ”O'F TO LEAVE A3? OUT. YORK 3L0”? FAIdLY

QUICKLY; DO NOT LABOR 0V3! AE'JY BLANK.

y L‘)\/‘

If we could devise openirw stratcnv

would wonderfully exemplify than: and aims of t*i~

‘l—IIU-I“~'-h?l“unfl ..2 1’.JJY‘-c>‘5‘.f\"

‘-

for our concern will be with a for sulancln" .13“

“WWW 3.173;!) t.‘- W.IL&M 3‘-

strategies,' ‘ i in contexts where the 'stvstngy’ itself

WM!“ {W‘kani‘ 'I

shall not in common use.

..MSI‘. W f)“;‘
D

The ' of interests to ours those in which peoylc

WW» . .:..'...:x.:$.z‘._:x?..:.r.~.r:

at cross purpc ,c:: in short, aituatign3. The pwablcm

dud-m. um ulna» 1:...- .14; mm

of ' to begin thcs apagcs ._ recognizablv of that typo;

“5th

'\
interests are Opposed -- in' polite way, of course,

Mina-Wu!“ - : -:.u:..).sq:a$ A

P

definitely opposed. For our to mos n you i.nto very

flammd‘auzw .c. L“! - : ..1. .5; i443.)

difficult type of activitW, while

0 -
'cu a osrson witn
wan-isms .

‘
J emu.“ .33. .9 4:1

enough troublca _ in may crave only relaxation satisfac tion

va’J-L? ,- A3. mzmuffiurg: .s:

of curiosity. This of interests are cancnti3.i the

Mtgi‘él’au cfiLLfi .5. . ‘ 4.131).?”

situations we shall '

o
“aways-max).

Another elcmc at is also and it is present upwln‘ ‘ too:

cum—w " 41¢ 1&3

Each of us- _ . cowgtcontrol over the . Many Q3 uiil accur

you: for ono,you throw the book at ' cat thus

' : * .. II IG""""D :‘ .3 :44 gvmarmw ‘

irritating both . _' writer and the cat, . at some cost in

perhaps some in self-respcc‘, undc vbtcCfl y Sore in
W

WM“, a.-. a.

deteriorated with the cat. 0? may skim the hard
‘ W '- W‘ -'

__‘h,irna,g and 80 on. There .- \._1da aSpsCtS within tpé control

.9“,

 

...a' 3.33.339':

aw.'.3..‘.¥3._-.." - C

the writer, too, such the choice and treatment - 'content ~-

but it is use css avy to labor the

. And a itrthnr
W5 .3413. 5:34‘25‘42. ru‘.

characteristics appea 3: Some npuct a oi ;4‘ particular

“Anal—Z... " ’ ...LIILW.‘

. v a , . .
.nd‘...‘. ___... . ..., 7‘ .,

situation is not t}se cantrs} of ¢‘_Ahp : ““3 I

-
ug-m.\:.r.--.an-. » - -. “_‘-.‘..:u-‘-c.u ‘ ' "

~¢. -.g

example, multittzda oi eve “.i; in p‘acu C‘“

aux—2&3.“ -- - .7 - : _v _. .—.'..A ..

 

‘ 11-“!1
I.

w

YOU AREIKKIASKED T0 INDICATE HOW YOU FEEL ABOUT THE ESSAY YOU HAVE JUST READ.
BENfiEN.mK}lPAlR OF WORDS YOU WILL FIND SEVEN BLAKKS. PUT AN X IN THE ONE

THAT REFLECTS YOUR REACTION. {FIRE IS AN EXAMPLE:

 

‘ :91 munc trszmzcu
COLORFUL __ __ ___ ___ __ _ .3... DULL
II’IIELLIGEEJT __ '______ ___: .35“ __ _ __ STUPID
won'mmss ___ ___ __ ___ _)5_ __ _ VALUABLE

If you felt that scientific research was very dull, you wnuld put an X in the
first line at the far right. if you felt thet it was neither intelligentrcr
stupid, you would put an X in the middle blank of the second line. And if you
felt that it was only slightly valuable, you would put an X in the third line

in the second blank from the right. Now put one X.in each line of the following
several lines, remembering that you are stnting your reactions to the essay you

have read.

THE NATURE OE STRATEGY

VALUABLE ' womzmsss

WEAK ___ __ ___ __ ___ __ __ ' 511mm
mmsnnr _' ___ __ __ __ __ __ UNPLEASANT
Hezmsz' __ __ ___. __ __ __ __ DISHONEST
PA 3 s 31w: __ __ __ __ __ _ ___ __ A51; IVE
UNE'AIR __ __ __ ___ __ ___. __ d/FAIR
’ LARGE, __ __ __ __ __ __ __ SMALL
CLEAR ___” __ __ __ __ __ ___ mzy
ROUGH ’ SMOOTH

BAD . GOOD

- Do you consider the quality of writing of the essay you have
just read to be good or bad? Indicate you answer by checking

the appropriate box below.
WELL WRITTEN

BADLY WRITTEN

Tunnz YOU VERY MUCH FOR YOUR COOPERATION IN THIS EXERCISE. WE
WILL TR? TO GIVE YOU THE RESULTS WITHIN A REASONABLE TIME AND
GIVE YOU OPPORTUNITY TO DISCUSS ANY QUESTIONS OR COMMENTS YOU

MAY HAVE THENO

55 Mi “IT! IN THESE BLANKB FILL IN THESE BLAUKS Version 4

 

Degree 7 Version

  

 

Student's N.me

 

V. A. W _
Class and Major

 

QESTRUCTIOFS

This in a _epeciel exercise which you are required to do during this class
period. The objective is to measure the group's ability to read text of the
kind that 1.. in your textbooks and to evaluate your ability to respond to it.
Although you will not be "graded" on the work you do during this exercise, a.
record will-'be kept of how well you do and some judgment will be made of your
ability. It will be prudent, therefore, for you to do as well as you can.

First you will be asked to read an essay of some 1250 wor de in length.

Read it carefully but do not try to memorize what you are reading. Read at

a speed which seems appropriate to you in terms of the text itself and your

own reading ability. Speed of reading in not a. matter of interest in this

exercise.

/

Next you will be given a new kind of test to see what you have been able

to‘ underntand from‘ the text. When you get to the rest, read its instructions

carefully; they are very brief but important..

‘ Third, andila'ot, you- will be asked to answer one question about the text,

your answer being one check mark in an apprOpriate Space. Again do precisely

what the instructions given there say.

As you go through the pages of this booklet, 99 N01" TURN BACKL ”_g PRI‘

CEDING PAGE. ,
. ‘ ’ I '
no nor ASK ANY QUESTIONS or THE msreucroas on or your. FELLQIL? srunmrrs.

None of then, probably, knows the answer. Read your instructions carefully

and you will know all you need to for this exercise.

TURN THE PAGE NW AND PROCEED. Read the essay and continue on through all

the pages of this booklet. when'you are finished, hold up your hand.

T2132 ll’fo’lE O}? STRATEGY

It is all too clear at this moment that there are many
ways for a book to begin; and most of those in plain sight
are transparently bad. We are tantalized by the thought that
somewhere among them g3“ lie hidden a few having such noble
qualities as these: The readers are informed -- perhaps with-
out suspecting it, though in the clearest prose -- of what the
writer intends to discuss; yet at the same time, it sounds like
the Lorelei calling. Hhercupon these readers resolve into two
groups: The first, a large and happy family really, will stick
to the book to the end, even though unimagined adversitieo imp
pend.‘ Further, this group will always think and speak kindly
of it, and will doubtless have at least one copy in every room.
The second group is most briefly described by stating that it
differs from the first; but the book acts immediately as a sopon
rific on all unpleasant passions, so as it is sleepily laid
aside, the sole lasting impression is that of a good gift sug-

gestion..

If we could devise an opeing strategy such as that, they
can wonderfully exemplify the theme and aims of the book, for

our concern throughout will he yith _a_ gethod for selecting best

i) A...

strateeies, even in contexts where the word 'strategy' itself
shall not be in common use.

The contexts of interest to ours are those inwwhich people
are at cross-purposes: in short, conflict situations. The

problem of how to begin these pages are recognizably of that type,

for certainly you and the writer are at cross-purposes, as our

 

interests are Opposed -- in a polite way, of course, but
definitely opposed. For our hope to ocean you into a very
difficult type of intellectual activity, while you a rea-
sonable person with enough troubles already, may crave only
relaxation or satisfaction of curiosity. This conflict of

interests are essential in the situations we shall study.

Another element is also essential and it is present
here too: Each of us exert £933 control over the situa-
tion. Many ways will occur to you: for one, you may throw
the hook at the cat, thus irritating both the writer am the
cat, hut at some cost in property, perhaps some in self-
r specs, and undoubtedly some in deteriorated relations with
the cat. Or you may ski the hard parts, and so on. There
are aspects within the control of the writer, too, such as
the choice and treatment of content -- but it is not necess-
ary to labor the point. And a further characteristics element
appears: Some aspects of this particular situation is not with— :k
in the control of either of you and I; for example, a multitude I
of events in our paste and extraneous influences during the ,
writing and reading periods played important roles. Of course U
~ these particular problem, of beginning the book in a real

optimum way, have a further characteristic which we shall hence-
. - /

forth shun, namely, it is too, hard -- else we should-had solved .

i2.

*4.

The restricti as on the subject matter be so fer-7 and

mild, it follow that the set of comlict situations we are
|

willing to consider is most notable for its catholicity.

o"he:.'e is no objects-too in p2.“ me Me, to consider an H~bozab

3
contest hem-men Here 3.13 M... ...1 0.2: a love affair of the
Barret ~Brotmiz eg tynz. 5‘40 (:0: st may be economic in

character, or it may be I-iusical Chairs. Or it may be almost

vitiee which ta}; 3 place éuring

w-

an}! one of the myriad ect
conventional war. it doesn't folio: that we had a nest-tum
for strategic ills in all these fields, but there is a

possibility that ours offering may as a method, perform

u-eful servic ein any of them.

The method which will be pxesented is identified by

" "he catch phrase Cate 313.92%. or, time permitting, the 3mg};

of Cams of Sit-232337;. if this were your first encounter with

QM . “cw-o. .II-o “ain‘t!“ --

that unlikelys use of nouns, the sole reaction is prob-

        
      

."
'.“"

'1.'M.~J:.'.I.-..--.::.- -

' ‘S- U'Orrpl'3"‘lfi“*

able: t-Ihy? Well, tl.e idea take their nmnes from the circum-

A

..A-l

stance that the study of games is a usefully and uoeble Gtart- *

V'ot "'4J't....'.,.AAA
"I A

‘00...

in: toint in the study of st:ate3fi. That does not really

61..
O

.-_—‘:1.-.. _.
g.

.. help, for ar- ,L. aim we hear: thy? Well, because games contains

' 9.1

0..
’11‘

tmny of the ingredients comma to all conflicts, and them

‘4‘! -
1+”.

.,. .009.
I

5. 3418?}-

are rela"ively amenable to description and to study. (In-

0

"g*‘w-4.~.,..

 

'9‘.

cidentally, having used the word 'game' to name the theory,

45;.

ib

' .

‘ ye then will call any conflict a game when our are considering

«...a-

them in the 15.3 ht of the theory. ) ,.-~
, , /

I.

,/

illustx ate the point, let us run our minds-over a

b. -

v t .
‘ ' A ' .4 ' c .
w" p.“ . , .

      

VOT‘.‘. ‘- L 4

T

5
3'...
oil--

‘

-IJO

Poker: game, keeping watch for items which are significant

cor

H40! Hove-r

Cir

in, say, a military conflict. You and four other are time

etc-wring human nature, undez. a 8373 otezn of r wards, your hope.

‘o-u ffga-OQO-
u A

Afi

' “CM.
' I. . .— ‘ v

We note at once that the players have opposing inter-
ests; each'want to win and, because tho'winnings of

one are necessarily the losses at another, their inter-
ests are oppose. This provides the bases of conflict.

We observe that some elements of the action, being e
matter of personal choice, is comnletely'within you
control. And the same being ttue for each player, there
are elements which are not within your control; worse,
they are controlled by minds having interests inimical to

yours. Finally, there are elements of the gems that are

~\ not,xunder the rules, within the control of any player,

such as the order of the cards in the deck. These elements
may be thought of as being controlled by Nature -- whom,*wc
believe, has a massively steble personality, a somewhat puck-
ish attitude toward your important affairs, but who beat you
no conscious malice. These are all surely familiar aspects

of any conflict situation.

Anothcr'charecteristic is that the state of informa-
tion -- intelligence, in the military sense -~ 18 a factor,
and, as usual, is an imperfect and hence troublesome factor;
We don't knOW'what the other fellow's hole card was. There
is also the bluff by which you, or the opposition, giues false
asidence regarding intentions\or strength of forces.//0ther

'

similarities will occur to you; peeple even get killed,

I

occasionally.

I

But the analog should not be pushed too far. You can

 

 

 

think of many aspects of warfare which is not reflected
in.Poher. One tank will sometimes kill two tents, in a
showdown; whereas a pair of Jacks always wins over an Ace-
high hand in the showdown. 'Of course Poker could be modi-
fied to make them contain showdown possibilities of this
kind, say by ruling that an.Ace is superior to any pair, up
to Jacks, whenever snybody's wife phones during the play of
a hand. But the fact is that the game don't exhibit all the
complexities of warfare and of other reel life-like conflict

situations ee which is precisely'why they are usable stertfi

 

\\ ing points for a study of strategy. In the early stages of

developing e theory it just is not possible simultaneously

to handle very many interacting factors.

-.P'

‘.“c I- - -
f-L'J'J'L' ' -

     
 

9 ‘0
31!. .

It is probably clear, then that games do contain some

 

.4. 2".

" - tu- ..
s‘ON u’ 0"" . ““

AJ-

     

'Woooli§1-

"' - - Ifo‘rlh.
“‘ “.-“finodonfl. no
a. - .-

of the basic elements that are present in almost any intet- i

‘0'.

I.

casting conflict situation. Does it follow that we can learn

.1». it, '

-— useful things by beginning a study'with them? Not necessar-
ily. ,It may be that sny nilitsry, economic, and social situa-

tion are just basically too complicated to be approached

 

through gene concepts. This possibility gains credence from

c the fact that the body of Game doctrine new in existence is

not even able to cope with fullublown real games; rather,‘we

are restrict at present to very simple real games, and to

T

rnxteredndown versions of complicated ones, such as Poker.

v. 'Q'ir- v‘—\ 7 'Ilj"‘\..""‘.‘- 7<~-y. 1.. r. . ' 1"- 9 :7“,:‘ ...,-2 'v‘"? "H"'r 7.1‘? I- . I -'~' ”7‘ T. "‘"I
n 21‘; .- EIJI C’.'\ 0 '..' l‘.,.,EJ, 9‘ LQ r' ..i' £I.‘IJ l).'.’-3.4A‘\.‘ ...1\ 3114 ‘;J/\0 3}) l‘l‘h ”1-5111 Lti'l 1l‘lk/‘A‘D
. 'c-"n‘ "man A -- \‘\ )3? u-\- .._,.. (’4'! m ’P 5"!) "“ 2,7,’\*\'r q - r .. -« ~A "\3' V
XI: 1’ Crull‘;"‘ "‘R """J' ~"}"-\‘-':~ ~l0. ["0 '12-'31,- lvtjl {O t'JJIA'H“ ‘~‘"‘ OL“ D \l\.‘l|\.‘ IA‘."J‘L.‘-. Eat..1“:l‘
I" f‘ “'9 "'..‘\ 'f‘fi T \‘_\f‘\7 i'l * v! .n
Qt--I\,:<i:1; n) I!“ l uh‘AJK Ovi-!\ .‘Il. .qu
~ 0 ‘u : ‘Q. .. , u's- '- us
I}. 110 9‘1). ’1' 3'13”; ‘3 013?;1111"; u ‘ j _ :Vv S‘ZCh <13 9
M0 r-— s-vr-c~rva ' m “"r. ‘3'Gr3"‘T-'W
L
" - I. 'u' . fl. ro- - 'n-O “ 3 I: l- ‘ :4.- " (‘I-l
LLXGZ’ C, .“ ‘-"d“k?-\‘4L‘J-Ul].‘r (31:... l"r“.c-""' ‘i’ L}lcr‘-Q 3.1(1 “$11.3 or .‘pr-fl' 5,. }J .l O‘ 9
. ' ' It”?,‘° v :V':~‘H'.'ffi r12 3"» " ‘ '
r 1:. w m: ' - mm nub-w
£01". 0". o- !‘ va-Qnrn 1‘) tn" ‘_ q:’_ Hp 01‘ t .2.- ‘, -.. Li) a t“, _j‘\| ‘ .u,.. .fl —.\'n-’
‘ h 1 .. “ ~ -4. ‘;“m-'Jntno " -1.“tll“-_' 2"..1”
.
. I . . _.‘ 4' ~. In.-‘.n f:
13-33? 53-351 ‘crflpl'; (33333 - Ck-‘T‘H. 3;:‘3 “(131135 tag V‘OYR! A .L'.(sbl..k
' ' " ' ' U ‘ 9 can; as: ,g . ..‘u ...zq-nv-maur-rs-ua
‘ . 3.
Snell no he ccmvsn use. _ 3
twtfimmfit’
I .3 .0 ‘ l
”3“ contnrta inte.;c to cure-ave in wagon
’ v ‘ ‘ . .3. “an... mam-a mm as",

’ O ‘ . P“ _
“cnla azu - crassvpu.r,csns: 1n snur‘;, can; Jict o the
' ‘ u.-.— — h—t— ‘ m

" ‘1

\

' ~ ~ - ‘0- C ’4 us 0 (s 'p. -‘
v." 1 ‘r 3 .r' - - )f'urh ncrr.) 03/309 A1." (3‘ tht K (V
Rd. Cb'fiiJ ‘ O). ’13-, wean: .H."¢-:n.l-Vr18 “‘4 .1 t‘ .' \ ‘ | ‘ mtu’kflv-fl'aab“ . ”J o

\ O

l ‘ ..., .... ‘ ,. ' . . 3 I. .‘ ‘ a .
rcr Yvd~ufid'tfiw wrbte wt Croas-pun‘ 03039 as our
mgr—sunsm’ .2 ta. mug-“Juana:-
. o. n '.
are cnooccd -- 1a a 'ay. of cc 3:33, but
ravsr‘x. ”.11: b flu) ° ‘ mm ran-“mummifi-
‘ o P Cf)! t .
gov-mad, Falw 0131M930 cczcn you late a dLLaLQflai

nflr—r'fi- “scum mfgm. .1.:r..rl“."'

typp of intallcatual ,rJ while you a '3rncn:hla . tha

I'm-man-rvxm. ... .

Ce‘.1C-11"‘11 trra'bls'u; slimy-16'” cravn cnly relaxation 0?

CW ’3' ‘ m:¢um

-‘ o o 0 . ,,_I.n .. , A ,— - . “ .
of ‘wvzcawfy. This confll t intern «Lo ere uSdetlu- 13

an M42“. :3“?! ‘ugquralranx: -3. M'-

clement is ciao es cntial it is prow3nt,h¢xm
.932:
: Eachwof us 3x33? ; cantrcl cvor tha situationo

vxyS‘will occur to z for one9 you may u" tha bcak at the

C38 Am‘m ' w 'T‘.w arm

(1'

I
a.) n r41.» :21: £3-.m=3

acme c353 in pr3p3 vty. - one in selfmre -..pe ct9 ali

“M ‘ ’ mam‘nmm

uzzdo *hau‘} 8432 in d3: epic 1~3tad with the cat” 0r

ma? skin the hard 9 and so cno- Thoma . aspects withiz

" '1. ‘. a ‘x. ...-r.:.-‘°..‘:: .3 .mtuLOJ'mxrrh-‘Ov-a
the cratrsl bra writerg tom, such thv choice and Lunat-

‘er'L‘ ‘3‘ n:.r.c. urn M m.” ‘11“ a
~~n~ ..- ‘&,~.| “. )‘u l,- .t .fi‘ . ‘ .n ‘1n.:.1o‘r \. x”? Q“ ti-”-
u .158. LL‘LS‘». 1‘: v ‘1» .- 1-.| " ‘ nC-VO‘ 33“") 'LO .. ‘“)OA 1'1
“MIMI-Wr‘tnr. u‘lmm
' '2. 'h“ '}' . "“- “- “u 0 9 ‘Of‘ a 'r‘n

‘ 0 mad a ti 3335101 c uI‘w.LOP13x.lCS appeal 3 . . ome “4.930....

“JM'.m _—

 

, has irri acing both t1w3M-and the catg but V . '

 

J -.

I I'
.-.,-1_

..
~ncrg-....,
' u. .

 

u can-a4.
.

3
Q:
I}
.I

i
a
‘2
I}

‘ I

..O

’ .. .
t'l'!..
-.I-fi— . p

YOU ARE FOE-4' ASKED TO 1.1‘70'LCATE HOW YOU I‘EEL ABOUT Tl‘iE ESSAY YOU HAVE JUST READ.

BE'WEEI-J EACH PAIR 071‘ WORDS YGU WILL FIND SEVEN BLANKS.

THAT REFLECTS YOUR REACTION. HERE IS AN EXAMPLE:

INTELLIGENT

WORTHLESS

[f you felt that scientific research was very dull,

RCIENTIFIC RESEARCH

 

 

 

 

PUT AN X IN THE ONE

DULL
STUPID

VALUABLE

you Wauld put an X in the

\ ‘.. .
First line at the far right. If you felt that it was neither intelligentrcr

Cupid, you would put an X in the middle blank of the second line. And if you

e1: that it was only slightly valuable, you wguld put an X in the third line

n the second blank from the right.

Now put one X.in each line of the follcwing

:verel lines, remembering that you are stating ycur reactions to the essay you

:ve read.

VALUABLE
WEAK
PLEASANT
110233. 81‘
PASSIVE
UNFAI R

' LARG
CLEAR
ROUGH

BAD

\

THE NATURE 9g STRATEGY

-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

WORN-{LESS
STRONG
UNPLEASANT
D ISIHONE ST

/
AgTIva

FAIR

SMALL

1} AZY

MOUTH

GOOD

\

 

u,.
- O.
- '-— I.
o
«pt-n . -

‘fl‘rr a.
o A, .
0'!

' ‘ .7: -
4" 4".'. J... ' ‘: ‘
I.

l
‘.' ‘n‘fl'uVflu’wfl

[at-‘9’.

‘ fl. '1 . U-O'( ;
o

-Do you consider the quality of writing of the essay you have
just read to be good or bad? Indicate you answer by checking

the appropriate box below.
WELL WRITTEN

BADLY WRITTEN

\

THANK YOU VERY MUCH FOR YOUR COOPERATION IN THIS EXERCISE. WE
WILL TRY TO GIVE YOU THE RESULTS WITHIN A REASONABLE TIME AND

GIVE YOU OPPORTUNITY TO DISCUSS ANY QUESTIONS OR COKHENTS YOU

MAY HAVE Tunno

 

 

APPENDIX B

COMPOSITION HANDBOOKS

-107-

 

COMPOSITION TEXTS SURVEYED

Barrett, Laurence. Writing for college. New York: American Book, 1956.

 

Bradford, Curtis, and Hazel Moritz. The Communication of ideas. Boston:
Heath, 1951.

Brennan, Maynard J., 0.S.B. Compact handbook of college composition.
Boston: Heath, 1964.

Buckler, William E., and William C. McAvoy. American college handbook.
New York: American Book, 1960.

Davidson, Donald. Concise American compoSition and rhetoric. New York:
Scribner's, 1964.

Gorrell, Robert M. and Charlton Laird. Modern English handbook.
Englewood Cliffs, New Jersey: Prentice-Hall, 1956.

Greever, Garland, and Basley S. Jones. The Century handbook of
’ Writing. New York: Appleton, Century, Crofts, 1942.

Guth, Hans P. Concise English handbook. San Francisco: Wadsworth, 1961.

Harrison, M. Clifford. Practical English grammar. Boston: Heath, 1947.

 

Hodges,_John C., Mary E. Whitten, and Francis X. Connolly. Harbrace
college handbook. New York: Harcourt, Brace, and World, 1962.

Ives, Sumner. A new handbook for writers. New York: Knopf, 1960.

Jones, Alexander B. Creative exposition. New York: Holt, 1957.

 

Kegel, Charles H., and Martin Stevens. Communication -- principles and
practice. San Francisco: Wadsworth, 1959.

Kierzek, John M. The Macmillan handbook of English. New York: Macmillan,
1949. ”

.-

Leggett, Glenn, C. David Mead, and William Charvat. Handbook for writers.
New York: Prentice-Hall, 1960.

Manchester, Frederick A. College English essential. New York: Appleton-
Century-Crofts, 1954.

McPeek, James A. S., and AuStin Wright. Handbook of English. New York:
Ronald, 1956. /

Myers, L. M. American English. New York: Prentice-Hall: 1952.
_ Ostrom, John. The craft of composition. New York: Holt, 1952.

Perrin, Porter“G. Writer's guide and index to English. Chicago: Scott,
Foresman, 1959.

Perrin, Porter 6., and George H. Smith. The Perrin-Smith handbook of
current English. New York: Scott, Foresman, 1955.

 

 

-109'

Phillips, Emma Julia. A review of English fundamentals. New York:
Holt, Rinehard, Winston, 1962. '

Rorabacher, Louise E. A concise guide to composition. New York:
Harper, 1963.

Shaw, Harry. Writing and rewriting. New York: Harper, 1951.

 

Suberman, Jack, and Henry M. Rosenberg. Basic Composition. Englewood
Cliffs, New Jersey: Prentice-Hall, 1956.

Taft, Kendall B., st 31., The technique of composition. New York:
Rinehart, 1960. '

Thompson, Wayne N. Fundamentals of communication. New York: McGraw-
Hill, 1957.

Waddell, Robert. Grammar and style. New York: Sloane, 1951.

‘Watkins, Floyd C., and Edwin T. Martin. Practical English handbook.
Boston: Houghton Mifflin, 1961.

Wilson, Harris W., and Louis G. Locke. The university handbook. New
York: Rinehart, 1960.

Woods, George B., and W. Arthur Turner. The Odyssey handbook and
guide to writing. New York: Odyssey Press, 1954.

Wooley, Edwin G., Franklin W. Scott, and Frederick Bracher. College
handbook of compgsition. Boston: Heath, 1958.

\

 

 

vav... -
tig..
,

“ :"-4uL-J: J1-...'.
9 1- .-~,...
..

‘i-n

 

I‘III 11111.!Ia ~11! .
.. . . . ......1. 3: (eat... . .5.» ,. .....l.
-. - . i. -. .. . : . .... ..u..a......ar...$w.n.a....s...c.........r.. . .... $3.31? NE... ......5 . ...-....:......... --.,- ...... ....
I IDDOI I ‘
1'IVOBC. .h 90...‘.0 ‘00.... It.‘ ’IJD‘IGO Val

 

.. .. ‘1‘! :5 u. [n’1:JT O NP‘!-rn...l$.‘m
Li.» ..r.....u....... In). .11”... 3r. ...-4.. . ... Sufi...» 1.21.59”: .

JIRWN‘OI'I} frnwo
. . . a
v0. ds'Kew-a”. In I.

. .....Ili ll.-
IIIIII.{.I.l¢a |:l-I\|.a111|.'l.ull.lllll..|l‘l.llllall III.
OI ...OII"OI$IO 1070!!
'0 n I 0.IoI'I.IIVCo-.|ltliiiul .. p
' .Ifi‘. IIIII .IOI...I
III. o I
...: -ritllu.,.ttlt'viiltt >105. QVQQDI‘o- -.I!l

"llllljlllllllllfll