E-Prime as a Tool for Critical Thinking

E-Prime! — The Fundamentals

Written by E-prime founder D. David Bourland, Jr.

Ambrose Bierce, in his famous Devil’s Dictionary, defined logic as “The art of thinking and reasoning in strict accordance with the limitations and incapacities of the human misunderstanding.” As we become conscious of our misunderstandings we improve the quality of our thinking, and most particularly our thinking about thinking, which Richard Paul defines as critical thinking.” In this article I will describe an offshoot of Korzybski’s system (18, 19) known as E-Prime: English without any form of the verb “to be.” The name comes from the equation E’ = E – e, where E represents the words of the English language, and e represents the inflected forms of “to be.”

We have heard of the unfortunate effects produced by the infamous “Person from Porlock” on English literature. A similarly unknown “Person from Tolland (Connecticut)” produced a more beneficial effect on me when, in 1949, I held a Fellowship for study with Alfred Korzybski at the Institute of General Semantics in Lime Rock, Connecticut. This Person, whose name I can no longer retrieve, wrote to the Institute suggesting that, in view of the problems Korzybski had discussed in connection with the “is of Identity” and the “is of Predication,” perhaps we should just abandon all uses of the verb “to be.” While no one else at the Institute seemed particularly interested in this suggestion, it struck me as having considerable merit, provided one could really do it.

The Institute staff at that time had perhaps reached its apex, consisting of Korzybski, M. Kendig, Charlotte Schuchardt (later Mrs. Allen Walker Read), Ralph Hamilton, Lynn Gates, myself, and David Levine. Guthrie Janssen had just finished editing Selections from Science and Sanity (20) and still lived in the immediate area. The letter from Tolland did not excite anyone else in this collection of critical thinkers, but it gnawed at me. The time fit me just right: I had a paper in preparation for the Third Congress on General Semantics. I decided to revise it once again to see if I could say, and indeed say better, what I wanted to convey without using any form of “to be.” In the process of this revision, I acquired an intermittent, but severe, headache which lasted for about a week. The final paper had a peculiar clarity, despite the fact that one “was” crept in. It eventually appeared as reference (2).

Between 1949 and 1964 I used E-Prime in several papers, but did not discuss this matter lest I become regarded as some kind of nut. Finally some close friends convinced me that a wider audience might feel interested in E-Prime. Encouraged particularly by my friend, Charles Chandler of Princeton, NJ, I prepared a brief report that appeared as reference (3). Subsequently two other studies discussed applications to the non-Aristotelian premises and to political documents. (4, 6)

In early 1969 the New York Society for General Semantics invited me to give a talk on E-Prime. This I did, trying to do so in E-Prime. By and large I succeeded (some people listened precisely for this — then and now they consist of a sharp, feisty bunch). Subsequently Kellogg (15) accomplished this and recommends it highly. His excellent ETC. paper contains many helpful suggestions.

Interest in E-Prime reached a local maximum in 1969. Due to mention of this oddity in a footnote of a paper by Allen Walker Read, Time magazine sent an interviewer to me to discuss E-Prime. A friendly article appeared in the issue of May 23, 1969. Unfortunately, in the process of writing this friendly piece, they attributed to me two quotations of statements I never made: one incredibly awkward, and the other such a gross transgression of E-Prime that it actually made publication at the bottom of a piece in The New Yorker magazine.

Let us turn now to what we can expect to happen to our writing and talking when we in fact do so in E-Prime.

Depending on exactly how one defines “word,” most scholars regard the English language as embracing some one to two million “words,” or lexical items. (Note 1.) In E-Prime one simply does without 20 or so of these lexical items; specifically, the “to be” family: be, is, am, are, was, were, been, being; plus contractions — ‘m, ‘s, ‘re; plus various archaic and dialectual forms — e.g., war, wert, beest, amn’t, ain’t.

While statistically E-Prime only makes trivial changes relative to the English lexicon, it does affect the syntax. Even this effect, however, does not seem as severe as it might appear. This unexpected lack of severity proceeds from the well-known “richness” of the English language, which provides a wealth of linking verbs (become, seem, appear, verbs related to the senses), apposition, etc., that can take over most of our habitual applications of “to be.” On the other hand, E-Prime does admittedly entail the necessity of expressing the progressive aspect by using continues to … and it makes use of the passive voice difficult or even impossible. (Note 2.)

In marked contrast with the areas of the lexicon and syntax, E-Prime delivers major and unexpected consequences to English semantics.

The E-Prime revision of English, although trivial in some respects, has deep underlying epistemological antecedents and consequences. Critical thinkers have struggled with the semantic consequences of the verb “to be” for hundreds of years. These distinguished persons include Thomas Hobbes (11), Augustus de Morgan (22), Bertrand Russell (24), Alfred North Whitehead (27) , George Santayana (25), and Alfred Korzybski (19).

Their concern, and ours as critical thinkers, centers upon two semantic usages of “to be,” Identity and Predication, that have these general structures:

(1) Noun Phrase-1 + TO BE + Noun Phrase-2 (Identity)

(2) Noun Phrase-1 + TO BE + Adjective Phrase-1 (Predication)

where TO BE represents an appropriately inflected form of the verb “to be.”

Critical thinkers have argued against using statements having the structure of (1) because they immediately produce high order abstractions that lead the user to premature judgments. Consider the following example:

(3) John is a farmer.

The immediate consequence of such an identification at the very least brings about unjustified abbreviation. For example, consider the following three sentences about “John”:

(4) John farms three acres.

(5) John owns and operates a 2,000-acre farm.

(6) John receives $20,000 a year from the government for not growing anything on his farm.

We could even carry this illustration into a different dimension:

(7) John, after living in the city all his life, has just bought a farm.

(8) John grew up on a farm and has farmed there for 61 years.

Despite the fact that (4) through (8) make extremely different statements about “John,” most English-speaking people feel comfortable making the jump from any one of (4) through (8) to (3). Critical thinkers trained in general semantics hold that (3) does not represent a valid higher order abstraction which could come from such observations as (4) through (8),but rather a possibly incorrect and certainly inadequate abbreviation of the larger picture.

Of course, due to the uniqueness of structures on the event level and the process character of “reality,” no structure can have precise identity with another — or even with itself at two different times, for that matter. Hence we can categorically deny the validity of any Identity relation. And accordingly, any linguistic structure which conveys or assumes an Identity relation does not correspond well with “reality.” As Korzybski would put it, “The map does not fit the territory.”

A decade before Korzybski, George Santayana described those matters somewhat poetically as follows:

The little word is has its tragedies: it names and identifies different things with the greatest innocence; and yet no two are ever identical, and if therein lies the charm of wedding them and calling them one, therein too lies the danger.

Whenever I use the word is, except in sheer tautology, I deeply misuse it; and when I discover my error, the world seems to fall asunder, and the members of my family no longer know one another. (25, p. 123.)

Let us now consider Predication, as illustrated in the following statements:

(9) The earth is flat.

(10) The earth is round (spherical).

(11) The earth is somewhat pear-shaped.

The verb “to be” carries with it a huge intellectual momentum of completeness, finality, and time independence. Still, each of the statements (9) through (11) does describe the earth adequately for some restricted purposes. This dual condition of adequacy-inadequacy seems characteristic of the Predication usage of “to be” and provides both its charm and danger.

Early presentations of Korzybski’s methodology evidently did not clearly explain the notion of the “is of Predication” despite its importance. Classical logicians have long called statements which have an “is of Predication” as the main relational term “subject-predicate” statements. As Bertrand Russell put it:

The belief or unconscious conviction that all propositions are of some subject-predicate form — in other words, that every fact consists of some thing having some quality — has rendered most philosophers incapable of giving any account of the world of science and dally life. (24, p. 24)

Korzybski stated the importance of this matter in the following way:

The subject-predicate form, the “is” of identity, and the elementalism of the Aristotelian system are perhaps the main semantic factors in need of revision, as they are found to be the foundation of the insufficiency of this system and represent the mechanism of semantic disturbances, making general adjustment and sanity impossible. (19, p. 371)

We may note in passing that the statements of both Russell and Korzybski contain one or more uses of the “is of Identity.” (See Note 3.)

We can agree, I trust, that the Identity and Predication uses of “to be” do not reflect factual circumstances in the world as we experience it. For those die-hards among us who have some doctrinaire bias, or who otherwise did not pay attention, I shall recapitulate:

  • Everything in the “real world” changes: sometimes so rapidly that we may not notice the changes directly (as in the case of a table which appears solid), sometimes so slowly that we can (as in the case of a river).
  • Every person, as well as every “thing,” undergoes such changes.
  • One particular verb in English — “to be” — carries with it archaic associations and implications of permanence and static existence that we do not find in the “real world.”

We have devoted much of the preceding material to a discussion of the epistemological reasons for avoiding the semantic usages of the “to be” of Identity and Predication. Other usages of that verb exist, of course, including the following:

(12) Auxiliary. (John is reading. Ivan is plotting. The rose is wilting.)

(13) Existence. (I am. Descartes was. You may be, but then again …)

(14) Location. (John is here. That is neither here nor there.)

I have heard that I. A. Richards has allegedly distinguished between some 23 different usages of “to be.” but I have never seen the paper in question.

For many years, as noted above, several titans of critical thinking have inveighed against the Identity and Predication usages, while continuing to use them. Piecemeal attempts to avoid the undesirable usages of “to be” simply have not worked. E-Prime provides a simple discipline that does. Even Korzybski and some of his most prominent students regularly fell into what we might call the “Is Trap.” I shall give three examples of the “Is Trap” in action:

  • Korzybski. Prior to the advent of E-Prime, Korzybski had more to say about the inherent dangers of the “to be” of Identity and Predication than any other critical thinker. And yet he himself fell into the “Is Trap” to the extent of using those two constructions in some 37% of his sentences in Science and Sanity.
  • Bois. For a number of years the late J. Samuel Bois served as the chief lecturer for the Institute of General Semantics at their annual seminars. Many, including this writer, consider his book, The Art of Awareness (1), an excellent introduction to general semantics. And yet Bois used the “to be” of Identity and Predication in about 42% of his sentences in that text.
  • Read. In a discussion of these matters, the noted lexicographer Allen Walker Read agreed that one should “call attention … to the undesirable ‘is of identity’ and ‘is of predication’ (as in reference (23)), but still rejected the most positive technique for doing more than just “call attention.” Read sought to justify his rejection on the basis of a set of allegations that do not apply to E-Prime (e.g., the latter does not make it impossible to express the progressive aspect, the passive voice, metaphor, adjectives, and appositives). He continues to use the “is of identity” and the “is of predication.”

Those three linguistically sensitive critical thinkers seemingly could not avoid the undesirable uses of “to be,” while allowing themselves the luxury of the other uses. At least, so they wrote — and spoke.

The Impact of E-Prime on Writing and Talking

In this part of the paper I will present four of the major consequences of using E-Prime in written and spoken utterances.

a. Vanishing Questions. One simply cannot ask a number of questions — some would say pseudo-questions — that have preoccupied many people. What is man? What is woman? Is it art? What is my destiny? Who am I? Such questions, by virtue of their semantic structure, set the stage for identifications and confusions in orders of abstraction. They tend to lead to discourse in which the likelihood of useful information generation or exchange declines precipitously. One might better ask questions on a lower order of abstraction such as these: What characterizes man or woman uniquely? In what way can I relate to this art form, if any) What can I do now to improve my future possibilities? May I have another drink?

b. Vanishing Internal Instructions. Various schools of psychotherapy have recognized the importance of the silent assumptions which we hold about the world and ourselves. Other schools, especially the “rational therapy” developed and practiced by Dr. Albert Ellis, also recognize the importance of what we tell ourselves, vocally and sub-vocally. “Self-suggested nonsense,” Dr. Ellis calls this in its undesirable forms. Most of us have encountered people whose life patterns have decayed as they keep repeating to themselves such comments as these: “I am a failure, consequently …” “I am a success, therefore …” “She is a Catholic, so …” “He is a Jew, hence …” “I am a teacher, so what I am doing must be teaching …” “Since I am the head of this household…”

c. Abbreviations. Forms of “to be” encourage and indeed facilitate the making of abbreviated statements that may turn out to convey little or no information, although we often behave as if they do. For example, we often see such empty comments as: “It is clear that …” “Well, business is business.” “The problem is just a matter of semantics.” Let us discuss that last assertion. While of course most human problems involve important (and usually unexplored because unperceived) semantic issues, these issues do not evaporate just because someone has labeled them thusly. Some people use “It’s just semantics” as an analysis stopper. One might productively respond to such a comment by pointing out, “Certainly; at least in part. Now let’s try to clarify some of those semantic problems.”

Confusion due to improperly abbreviating with “to be” even occurs in primary schools. All too frequently we still hear teachers insisting that children drill in arithmetic by saying “One plus one is two; one plus two is three; etc.” The perfectly correct mathematical expression “equals” certainly need have no more inherent mystery for the young than “plus.” The unnecessary use of is in this context may have some responsibility for the difficulties some children experience with fractions. They can readily see the differences between 1/3 and 2/6, say. The first fraction may equal the second, but obviously some trouble could arise for those taught to translate as ” = ” as “is.”

d. Return of the Role Players. As mentioned above, E-Prime makes use of the passive voice somewhat difficult. One may have to resort to constructions with the somewhat scruffy auxiliary verb “to get” as in “The work got done.” Rather than a drawback, this consists of one of the greatest contributions of E-Prime. This facet of E-Prime forces users to bring the role players into explicit prominence or to indicate their ignorance of them. For example, many writers of technical and scientific papers forget that objectivity resides in the persons conducting the various experiments, etc., rather than in the passive forms used in reporting the results. I know of two instances in which scientists applied E-Prime to their complete report because this technique actually forced them to make explicit some important early details. One instance involved the failure of a sensor on a satellite, and the other concerned the fact that contractor personnel did not switch on a certain antenna. In both instances early versions of the reports in question said something like, “The data were not available.” Subsequent digging for the role players brought useful information to light.

Politics and Language

In the years immediately following World War I, Alfred Korzybski observed the stark differences between the consequences of engineering and scientific activity and the fruits of political activity. He pointed out that, when engineers build a bridge it normally functions as designed. But when politicians “build” a treaty or government, it usually collapses amid great human suffering.

Korzybski’s analysis led him to conclude that the fundamental factor responsible for that discrepancy in performance consists of the structure of the languages used by those who design bridges and those who design governments. The engineers and scientists use a language (mathematics) which has a structure similar to that of the bridges, hence the language produces predictability. However, the politicians normally employ a language of archaic structure that uses static terminology in describing dynamic human socioeconomic issues. As Korzybski pointed out, to the extent that a treaty, constitution, etc., incorporates this kind of static-dynamic discrepancy, one may expect undesirable and unstable consequences.

To put this somewhat differently, Korzybski asserted in his books that dynamic social institutions, if based upon static premises, must ultimately collapse. And if we inquire into this matter semantically, we find that the use of the verb “to be” constitutes the main source of static premises and assertions in ordinary English.

Recognizing the insidious role which “to be” theoretically may perform in socio-political contexts, I analyzed several important, basic political documents. The purpose consisted of determining to what extent the language in the documents exhibited a static character, as indicated by their reliance on the “is of Identity” and “is of Predication.”

I chose the following political documents for study:

a. The Constitution of the United States of America.

b. The Communist Manifesto.

c. Machiavelli’s The Prince.

d. Robert Welch’s The Blue Book.

e. Aristotle’s Politics.

                               TABLE I
                        POLITICAL DOCUMENT STUDY

                                       Per cent of Sentences
                      Sentences        With One or More Uses of
Document              in Sample        Identification or Predication
Constitution of U.S.A.
 a. Main Body a.          99 b.                20.2
 b. Complete             166 b.                21.6

Communist Manifesto      444 b.                26.2
The Blue Book            207                   48.8
The Prince               175                   53.6
Politics                 188                   60.1
Notes: a. Exclusive of amendments.
       b. Complete document, not a sample.

Table I shows the results of the analysis of sentences in the documents noted above. Some distortion in the results may exist, due to the fact that the documents by Machiavelli and Aristotle exist as English translations. However, the original languages in both cases belong to the Indo-European family, so the distortion probably does not amount to too much. Marx and Engels allegedly wrote the “Communist Manifesto” in several “original” languages, including English. Engels supposedly edited the English version which I analyzed.

In my assessment, the results given in Table 1, ranked in accordance with the increasing appearance of the uses of Identity and Predication, also correlate precisely with the great flexibility and power of our Constitution to the sterility of Mr. Welch’s nightmare, and the rigid dogmatism of Aristotle. I submit that these results give quantitative substantiation for Korzybski’s thesis.


Apart from any doctrinaire considerations, E-Prime can assist the user in attaining a kind of vigorous clarity that many have found worthwhile. Of course I know of only some of the people who have found E-Prime useful in their writing and speaking. However, E-Prime has found application in: one doctoral dissertation in physics (by Dr. D. A. Schwartz in 1968) (26), one licenciatura thesis in Linguistics (mine in 1973) (5), a master’s thesis in Biblical studies (by Byron L. Cannon in 1987) (7 ), a multi-volume research report by the U.S. Naval Air Systems Center (Project IMP in 1971), and a variety of papers published by myself, E. W. Kellogg, III, Elaine C. Johnson and Paul Dennithorne Johnston in the General Semantics Bulletin and Et cetera.

Of course, it pleased me greatly to learn that the noted psychotherapist, Dr. Albert Ellis, thought enough of the benefits of E-Prime that he re-wrote two of his books in this manner (A New Guide to Rational Living, with Robert A. Harper in 1975, and Anger: How to Live With and Without It in 1977). (9, 10) Scientific papers by Kellogg which show “E-Prime in action” have appeared in Nature (17), The Journal of Bioelectricity (14), and The Journal of Gerontology(16). Further applications have appeared in various places due to the efforts of C. A. Hilgartner, M. D., K. L. Ruskin, M. D., Charles Morgan, and T. J. Hefferon.

The diversity of applications described above testifies to the generality and utility of E-Prime.

I offer E-Prime to those interested in critical thinking as an easily teachable technique that has immediate benefits. In writing and talking it provides a method for materially reducing “the human misunderstanding.” As the current ad for Nike Athletic Shoes puts it, “Just do it.”


1. Bois. J. Samuel. The Art of Awareness. Dubuque: Wm. C. Brown Co. 1966.

2. Bourland, D. David, Jr. “Introduction to a Structural Calculus: A Postulational Statement of Alfred Korzybski’s non-Aristotelian Linguistic System,” General Semantics Bulletin Nos. 8 & 9. 1952.

3. Bourland, D. David, Jr. “A Linguistic Note: Writing in E-Prime,” General Semantics Bulletin Nos. 32 & 33. 1965.

4. Bourland, D. David, Jr. “The Semantics of a non-Aristotelian Language,” General Semantics Bulletin, No. 35. 1968.

5. Bourland, D. David, Jr. A non-Aristotelian Paradigm for Linguistics. Licenciatura thesis. San Jose, Costa Rica: Universidad de Costa Rica.

6. Bourland, D. David, Jr. “The Language of E-Prime,” in Coping with Increasing Complexity, Ed. by D. E. Washburn and D. R. Smith. New York: Gordon & Breach. 1974.

7. Cannon, Bryon L. An E-Prime Approach to the Holy Bible. MS thesis. Fort Hays, Kansas: Fort Hays State University. 1987.

8. Chomsky, Noam. Syntactic Structures. The Hague: Mouton & Co. 1957.

9. Ellis, Albert. Anger: How to Live With and Without It. Secaucus, NJ.: Citadel Press. 1977.

10. Ellis, Albert., and Robert A. Harper. A New Guide to Rational Living. North Hollywood: Wilshire Book Co. 1975.

11. Hobbes, Thomas. Leviathan. 1651. (New York: Liberal Arts Press edition 1958).

12. Jesperson, Otto. Essentials of English Grammar. London: Allen & Unwin. 1933.

13. Johnson, Elaine C. “Discovering E-Prime,” Et cetera: A Review of General Semantics, Vol. 45, No. 2. 1988.

14. Kellogg, E. W., III. “Air Ions: Their Possible Biological Significance and Effects,” J. Bioelectricity, Vol. 3, p. 119-136. 1984.

15. Kellogg, E. W., III. “Speaking in E-Prime, “Et cetera: A Review of General Semantics, Vol. 44, No. 2, 1987.

16. Kellogg, E. W., III., and M. C. Yost. “The Effects of Long-Term Air Ion and D. C. Electric Field Exposures on Survival Characteristics in Female NAMRU Mice,” J. Gerontology, Vol. 41, p. 147-153. 1986.

17. Kellogg, E. W., III., M. C. Yost, and A. P. Kruger. “Superoxide Involvement in the Bactericidal Effects of Negative Air Ions on Staphlococcus albus,” Nature, Vol. 28 1, p. 400-401. 1986.

18. Korzybski, Alfred. Manhood of humanity. New York: E. P. Dutton. 1921. Second edition, 1950, Distributed by the Institute of General Semantics.

19. Korzybski, Alfred. Science and Sanity. Lakeville, CT: International Non-Aristotelian Publishing Co. 1933. Fourth edition 1958.

20. Korzybski, Alfred. Selections from Science and Sanity. Lakeville, CT: International Non-Aristotelian Publishing Co. 1948. Eighth Printing 1976.

21. Lee, Laura L. “The Relevance of General Semantics to the Development of Sentence Structure in Children’s Language,” in Communication: General Semantics Perspectives, Lee Thayer, ed. New York: Spartan. 1970.

22. de Morgan, Augustus. Formal Logic. 1847. London: Open Court (reprinted 1926).

23. Read, Allen Walker. Personal Communication. 1976.

24. Russell, Bertrand. Our Knowledge of the external World. Chicago: University of Chicago Press. 1914.

25. Santayana, George. Scepticism and Animal Faith. New York: Scribner.

26. Schwartz, Daniel A. The Spatial Distribution of the Diffuse Com ponent of Cosmic X-Rays. Doctoral dissertation. San Diego: University of California at San Diego. 1968.

27. Whitehead, Alfred North. The Principle of Relativity With Applications to Physical Science. Cambridge, England: Cambridge University Press. 1922.


1. This conventional position ignores the names of the integers after some arbitrary cut-off point. Other-wise, we would have to say trivially that most modern languages contain at least a denumerably infinite number of words.

2. The comparatively minor syntactic consequences of E-Prime reflect the same linguistic functions operating that account for the fact that some natural languages lack a verb that corresponds exactly to “to be,” including Russian, Hungarian, and Mandarin at least. It seems interesting to note that the speakers of those languages alone account for about 20% of the world’s population.

3. The lack of an adequate treatment of the “Is of Predication” has led to some unfortunate confusion. The “Is of Predication” statement has the basic structure given in (2) above. In the early days of the transformational approach to syntax, Chomsky (8 ) gave the following re-write rule for a sentence (S):
(N 1) S —> NP + VP,
where NP represents noun phrase, and VP represents verb phrase. In pre-Bloomfieldian “traditional” grammars linguists called the NP of (N1) the “subject,” and the VP of (N1) the “predicate” (e.g., Jesperson (12, p. 97). The slight difference in terminology and the great difference in significance between the philosophical subject-predicate and the linguistic subject + predicate provided the raw material for problems. For more on this matter see reference (21, p. 121).

A retired associate professor of Linguistics, D. David Bourland, Jr. has studied and written in the field of general semantics for many years.

First published in ETC: A Review of General Semantics, Vol. 46, No. 3, Fall 1989. Copyright © 1989 by the International Society for General Semantics.

“Person from Perlock.” This refers to Samuel Taylor Coleridge and his writing of the poem Kubla Khan early in the 19th century. The story says he’d dreamed a much longer piece while in an opiate trance. As he wrote it down, the infamous “person from Porlock” interrupted him on trivial business and made him forget much of what he’d intended to inscribe. [Ed]

pc updated 10/20/95

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.