Trinity Paper No.

- LANGUAGE
AND
THEOLOGY »
 Digitized by the Internet Archive
in 2024

https://archive.org/details/languagetheology0000clar
LANGUAGE AND THEOLOGY

Gordon H. Clark

PRESBYTERIAN AND REFORMED PUBLISHING CO.

Phillipsburg, New Jersey
LEER 1980
San Diego Christian Couege
i Forevy
Gartana (“A
 Copyright 1979
The Trinity Foundation

ISBN: 0-87552-141-X

PRINTED IN THE UNITED STATES OF AMERICA

 Table of Contents

Ore OTe es or Cee ee ee ek eet Vv

PART ONE: SECULAR THEORIES

Pee rrerss tee OP LAN CO ae oe ees Vetek oS Po dee 3
RIPLEY 41EOE Cs cea cee nau neuen voc awe 5
PRICES A ICIISSONL, Oe oa ea wishes Mie UP aaa acdiecs 8
a. Difficulties in Understanding Him ................. 8
Dre ee VATD cok ies Ee i ee ane citi ancet 10
in ine meed of a New Langage oos6ce cee oss darcien « 10
ii. The Difficulties of Ordinary Language .......... 12
SENECA ONE NN Bois yoo asus ns dais Morncdize Eps DAS)
BOM CPG PAT RET gros rata a i a eR ga 36
er Eee a aed ie alana ala Aseianicd'S hap ulhae 43
eC GE I OU e eee a e Sit Nadas Worms aaa Was 54
OE PR Pee 7c de a 63

PART Two: RELIGIOUS THEORIES

as WYERIN TO ORT OsRAN oe as > a slSts analy Pca Hotere oral Ss 85
TAN Ge aoe is casos sara iicecai'd ois, atuoice Save GOS 95
PET OU ACE RVRIG TRICE is Gees es bata Poo alaue asian d a Gama 102
a AN CUON a Oy, oe ateetnds are Seeks as Peo Raat v 105
Hem LTCeE Ce rot ee ae ne Ui boom ture see Pow eras 111
Wal ATHELOrGeR eros tus ee the nan decwnee aioe 115
TM GHUCUNY HAITI fines os Go he aes eee ae Nae 122
SeAPCHristan CONStLUC LION cos it wna $0.5 fines basels sg 688 Bal

ili
 Foreword

HISTORIANS have christened the thirteenth century the Age of

Faith and termed the eighteenth century the Age of Reason. The
twentieth century has been called many things: the Atomic Age,
the Age of Inflation, the Age of Envy, the Age of Aquarius. But
this century deserves one name more than any other: the Age of
Irrationalism.
The Trinity Foundation believes that there is no greater threat
facing the true Church of Christ at this moment than the irra-
tionalism that controls our entire culture. That irrationalism is so
pervasive that even the Remnant—the segment of the professing
church that remains faithful—has accepted much of it, frequently
without even being aware of what it was accepting. This book
examines and refutes one aspect of that irrationalism, the notion
that human language is an inadequate tool for the communication
of divine truth. The attack of the irrationalists on revelation has
never been more subtle than their attack on language, for it
allows them to posture as defenders of an inexpressible divine
“truth,” while destroying the means by which that truth is com-
municated to men.
The purpose of the Trinity Foundation is to expose and combat
the irrationalism of the present age, whether its current spokes-
man be an existentialist philosopher, a professed Reformed
theologian, or a Roman Catholic charismatic. To accomplish this
purpose, the Foundation has undertaken a program of publishing
works that present the system of doctrine taught in the Scriptures
as clearly and unambiguously as possible. We do not regard
obscurity as a virtue, nor confusion as a sign of spirituality.

Vv
vi FOREWORD

The Trinity Papers are a series of books that expound the ideas
without which Christianity must disappear: the Bible as the sole
source of truth, the primacy of the intellect, the supreme impor-
tance of correct doctrine, and the necessity for systematic and
logical thinking. This volume is the first of those Papers. The
Trinity Foundation also publishes The Trinity Review, an occa-
sional newsletter of the Foundation. Its distribution is free and is
available to anyone who personally requests it. The Trinity Mani-
festo is the statement of purpose of the Foundation. Free copies of
either the Review or the Manifesto may be obtained from the
Foundation at Post Office Box 169, Jefferson, Maryland 21755.
Additional copies of this volume are also available for purchase
from the Foundation.
JOHN W. ROBBINS
President
 PART ONE:
SECULAR THEORIES
 AO THAT 7
2UROANT AAMAS
— a
 1
Scope and Importance

MANY religious writers today deny the adequacy of human

language to express truth concerning God. A statement that God
is omnipotent, or merciful, or even that he directed the exodus of
the Jews from Egypt, is taken as mythological, parabolic, analog-
ical, or as some poorly defined “pointer” to an unknowable realm
beyond human comprehension. Such theories of religious lan-
guage are partially a reaction against and partially an adaptation
of secular theories that had previously disposed of all theology
and metaphysics as utter nonsense. Therefore, to understand the
religious theories, one must have some knowledge of their secular
predecessors.
This language philosophy, whether in its most extreme form of
logical positivism or logical empiricism, or in a less extreme form
of the philosophy of analysis, or in the later forms of ordinary
language philosophy, is astrictly twentieth century phenomenon.
Of course, earlier philosophers showed some interest in lan-
guage. Plato lampoons the pedantic purism of Prodicus and takes
note of more serious questions in his dialogue Kratylus. Aristotle
is praised or more commonly berated for finding physical and
metaphysical guidance in grammar. The Stoics, too, not only
pursued the details of grammar and argued for their fatalism on
the basis that some propositions are true; they also considered
how a word could refer to a thing, and so produced a theory of
signification which conflicted with their own basic materialism.!
Finally Augustine of Hippo wrote a famous treatise, De Magistro,

1Cf, Emile Bréhier, La Théorie des Incorporels dans l’'Ancien Stoicisme.

3
4 SCOPE AND IMPORTANCE

on the relation of language to thought and how a pupil could learn

from a teacher. But none of these ancient theories of language
matched the modern language schools in their intricate detail,
nor in their sweeping conclusions about theological language.”

2There is acertain inaccuracy or inadequacy in these statements. Not to mention

Rousseau, who dropped the problem in despair, there were also K. F. Becker,
Wilhelm von Humboldt, Fichte, Herder, George Smith, Josiah Gibbs, Max Miller,
Benjamin J. Taylor, Charles de Brosses, and Horace Bushnell.
 2
Skeletal History

BEFORE the sometimes trivial and often puzzling details are

studied, and before the substantial positions and arguments are
weighed, a skeletal history of the general development might
prove of some smail value. It will be hardly more than a list of
names and dates. The dates themselves are not so important as
dates are in other matters because these men are roughly con-
temporaries and a strict chronological listing of the theories
would obscure any logical continuity. Even so a list of names can
provide a convenient reference.
For somewhat insufficient reasons G. E. Moore (1873-1958) is
often considered as the initiator of language philosophy. The best
reason is his influence on Bertrand Russell. Moore and Russell
were originally Hegelians and disciples of Bradley. Between
1898 and 1904, during frequent discussions with Russell, Moore
turned himself and Russell away from Hegelianism, the result of
which was his article The Refutation of Idealism in 19038. After
1911 he began to think of language. Now in view of Plato and
Augustine there seems to be no reason why thinking of language
should make a person an empiricist. But the twentieth century
movement, following the realism of Moore, has been almost
unanimously empirical. However, though Moore may have been
astimulusor an initiator, he can hardly be called the founder of a
sehool—and that for two reasons: (1) there really was no single
school, and (2) Moore, by 1940, after various peregrinations,
seems to have returned to something like Bradley again.
It is better to identify Bertrand Russell (1872-1970) as the
founder of language philosophy. To say how, at this point, would
take us far beyond any skeleta! history.
5
6 SKELETAL HISTORY

Following Russell in time, but at first independent of him, was

a group that can more properly be called a school of philosophy.
This was the “Vienna Circle” of logical positivism. Despite some
loose language in the historical accounts of twentieth century
philosophy, Bertrand Russell was not a logical positivist, though
their interests overlapped. The “Vienna Circle” was a group of
thinkers who were particularly interested in the philosophy of
science. These men were voluminous publishers. In 1917 Moritz
Schlick, who may be called the organizer of the school—he died in
1936—wrote Space and Time in Contemporary Physics. In 1919
Ludwig Wittgenstein, whom Neurath and some others considered
not quite orthodox, produced his famous Tractatus Logico-
Philosophicus, the doctrines of which he later repudiated as he
drew still further away from Vienna orthodoxy.
Rudolf Carnap published The Elimination of Metaphysics
through Logical Analysis of Language in 1932, and in 1934, The
Logical Syntax of Language. Then there is Herbert Feigl, who
later settled at the University of Minnesota, the author of The
Logical Character of the Principle of Interaction (1934), Scientific
Method without Metaphysical Presuppositions (1954), and much
else.
Not only did the Vienna men publish books, but also in 1929
they organized an international congress, held at Prague; and
other congresses followed. One may also mention Otto Neurath,
Philipp Frank, and Kurt Godel. These men were driven from the
comfort of their Vienna home by the increasing power of Hitler,
and found refuge chiefly in England and America.
With the geographical dissolution of the Vienna Circle, the
term logical positivism came to be extended to several philos-
ophers who, though sympathetic with the main ideas of the
Vienna group, were not so strictly scientific and who also began
to diverge in various directions.
A.J. Ayer made a great impression in 1936 with his Language,
Truth, and Logic; a second edition, with a modifying introduction,
came ten years later.
Along with Ayer’s first edition, Gilbert Ryle in his Systematic
Misleading Expressions (1931) and his Concept of Mind (1949) is
doubtless closer to the original Circle than most of the later
SKELETAL HISTORY 7

authors. There is no Circle now; but language philosophers, espe-

cially religious philosophers, flourish in abundance.
Such is a skeletal history of language philosophy, especially
secular language philosophy. Its most vigorous anti-religious and
anti-metaphysical exemplars were the logical positivists. But
even in their heyday, they did not succeed in convincing every
language philosopher. There were some, not necessarily orthodox
in theology, who refused to dismiss religious language as syntac-
tical nonsense. Much less would they abandon poetry and moral-
ity. Wilbur Marshall Urban in Language and Reality (1924) may
be conveniently taken as a logical, though not a chronological,
intermediary between the uncompromising anti-religious logical
positivists and the religious writers of a later date. The present
program therefore will begin with Bertrand Russell. Then the
logical positivists can be interpreted as more consistent exponents
of some of his views. Carnap surely expressed the anti-metaphysical
conclusions of language philosophy more clearly than Russell
did. Indeed, Russell himself was not anti-metaphysical, however
anti-religious he may have been. Following this will come A. J.
Ayer and the difficulties of formulating a satisfactory principle
of verification. After this the account must reverse the chronology
and return to Urban in 1924. And finally there will be the explicit,
though diverse, defenses of some sort of religious language.
 3
Bertrand Russell (1872-1970)

a. DIFFICULTIES IN UNDERSTANDING HIM

ALTHOUGH Bertrand Russell, both by chronology and by reason of

the less consistent content, is more properly the founder of the
modern language philosophy than the Vienna group, and should
therefore receive extensive consideration, it is almost impossible
to give an account, both accurate and complete, of his position.
The first reason is that his contributions are voluminous and
detailed, very detailed, with the result that any complete account
would be more voluminous than Russell himself. The second
reason is that he changed his opinions too frequently. This fact
may do honor to his honesty in acknowledging flaws and correct-
ing them to meet objections, often his own; but it imposes on
critics the burden of writing a chapter or a book on Russell I, and
a second volume on Russell II, on to Russell IX or so. The third
and most annoying reason is that by hisown admission he has not
said and cannot say what he means. In reply to Bradley, with
reference to unities and simples, Russell says that “the topic is one
with which language by its very nature is peculiarly unfitted to
deal. I must beg the reader therefore to be indulgent if what I say
is not exactly what I mean, and try to see what I mean in spite of
unavoidable linguistic obstacles to clear expressions.”! Again,
this does honor to his honesty, but it also suggests that perhaps he
might just have stumbled into an impossible task.

‘Bertrand Russell, “Logical Atomism” (1924), in Logical Positivism, ed. by A.J.

Ayer (New York: The Free Press, 1959), p. 43.

8
BERTRAND RUSSELL 9

In spite of many modest assertions of the tentativeness of his

opinions, Russell is bold in “endeavoring all the way through to
make the views I advocate result inevitably from absolutely
undeniable data.”? Then he shortly adds, “the data which are
undeniable to start with are always rather vague and ambiguous.”
How can anything result inevitably from what is vague and
ambiguous? On the very next page he acknowledges that “when
you pass from the vague to the precise by the method .. . that I
am speaking of, you alwaysrun acertain risk oferror. . . . Ishall
run a great many risks and it will be extremely likely that any
precise statement I make will be something not true at all”
(p. 180).
Other passages also show that Russell does not understand the
meaning of the words he writes; and on his own assurances it is
certain that no one else knows what Russell means. “When one
person uses a word, he does not mean by it the same thing as
another person means by it... . It would be absolutely fatal if
people meant the same thing by their words... . The meaning
you attach to your words must depend on the nature of the objects
you are acquainted with, and since different people are acquainted
with different objects, they would not be able to talk to each other
unless they attached quite different meanings to their words”
(ibid., p. 195). Note that Russell does not say that sometimes
people use the word in different senses; he says (implicitly)
always. Perhaps he thinks it an undeniable datum that no two
people ever use a word in the same sense. But it is at least possible
that one or two persons might be bold enough to deny it. At any
rate, on Russell’s own principles, no one who discusses him can
know what he meant.
Perhaps Russell knew what he meant, sometimes; but on very
fundamental matters he quite often did not know what he was
saying. After laying down certain “provisional definitions,” he
added, “This is not absolutely correct, but it will enable you to
understand my meaning” (ibid., p. 196). “All our words are
ambiguous” (ibid., p. 197).

“The Philosophy of Logical Atomism” (1918), in Logic and Knowledge, ed. by R.

C. Marsh (London: George Allen and Unwin Ltd., 1956), pp. 178-179.
10 BERTRAND RUSSELL

There is also a slightly different type of difficulty that a critic

must face. Russell sometimes makes statements that he retracts
pages later. Then the critic has to reread the earlier pages or
chapters to see if the later assertion alters the force of the inter-
vening argument. This may not be an impossible task for the
critic, but it is annoying. For example, “But a belief is true or
false in the same way a proposition is, so that you do have facts
[this belief] in the world that are true or false. I said a while back
that there was no distinction of true or false among facts, but as
regard that special class of facts that we call ‘beliefs,’ there is”
(ibid., p. 227).
Or, again, “I have been talking, for brevity’s sake, as if there
really were all these different sorts of things. Of course, this is
nonsense” (2bid., p. 265).
Kindly permit one further and final reference: “In some
respects, my published work, outside mathematical logic, does
not at all completely represent my beliefs or my general outlook.”
Taken very strictly, these references make the work of a critic
as impossible as it is useless. This may seem to be an extreme and
ungrateful conclusion; but Max Black, in the Schilpp volume
(pp. 229-231), says much the same thing, even if in gentler terms.
However, Max Black and the present writer cheerfully acknowl-
edge that Russel I to Russell IX can supply some valuable
thoughts, not to the critic or historian as such, but to the construc-
tive thinker for the development of his own position, provided
always that the constructive thinker resolutely rejects Russell’s
proposition that no two people ever use a word in the same sense.

b. EXPOSITION:
i. THE NEED OF A NEW LANGUAGE
The exposition of Russell’s philosophy, for the present purpose,
ought to stay close to his views on language. But any theory of
language soon merges with psychology, logic, and perhaps

’The Philosophy of Bertrand Russell, ed. by Paul A. Schilpp (New York: Tudor
Publishing Co., 1944), p. 16.
BERTRAND RUSSELL ial

metaphysics, not to mention epistemology. These subjects, of

course, use language, and most philosophers write in English,
French, German, or Greek. Aristotle’s Greek was somewhat
technical, Plato’s was ‘ordinary language,’ however literary in
style. But Russell finds that ordinary language is so confused and
confusing that its effects on philosophy have been disastrous; and
therefore a new language is necessary.
For example, Russell says, “It is exceedingly difficult to make
this point clear as long as one adheres to ordinary language,
because ordinary language is rooted in a certain feeling about
logic, a certain feeling that our primeval ancestors had, and as
long as you keep to ordinary language[which Russell on occasion
can use with tremendous literary effect] you find it very difficult
to get away from the bias which is imposed upon you by
language.”4
This long sentence alludes to the psychology of primitive
peoples, to the effect of their bias on the origin of language, and,
in the context, to the relation of propositional functions to propo-
sitions and individuals. At the moment it must serve only to show
Russell’s rejection of ordinary language and his desire to create
an artificial ideal language. He enforces this point in the following
paragraph by saying, “I think an almost unbelievable amount of
false philosophy has arisen through not realizing what ‘existence’
means.” And between these two sentences he says, “The only way
you can really state it correctly is by inventing a new language ad
oe ae
Again, in Logical Atomism (ed. by A. J. Ayer, p. 36) Russell
says, “The ontological argument and most of its refutations are
found to depend on bad grammar” (See Principia Mathematica,
p. 14). Of course, he does not mean that Anselm used split infini-
tives or that Kant failed to make adjectives agree in gender:
grammar for Russell is something more than grammar, for
unfortunately the words of ordinary language are not precise. In
other places he shows how confused it is to say, “Scott was the
author of Waverly,” and “all men are mortal.”
This is sufficient to show that Russell wanted to invent an

‘The Philosophy of Logical Atomism, ed. by Marsh, p. 234.

12 BERTRAND RUSSELL

artificial language and so escape the paradoxes of the mother

tongue.

ii. THE DIFFICULTIES OF ORDINARY LANGUAGE

These assertions of the need for anew, artificial, ideal language

introduce us to the substance of the argument. The first question
must be, What, specifically, are the paradoxes that need solution?
Everyone recognizes that language and expression are beset
with ambiguities and misunderstandings. This is an everyday
occurrence. But it does not follow that every such instance is the
key, or the lock, toa profound metaphysical blunder. It is at least
possible, subject to further examination, that some of the difficul-
ties Russell alleges are quite trivial. Others of them have long
been recognized as both difficult and embarrassing. Anyone who
has studied a little medieval philosophy, or is to some extent
acquainted with the history of logic, knows about the insolubilia—
problems so difficult that no solution is possible. Now, Russell
believes that they can be solved. But the critic must ask, Cannot
they be solved by ordinary language? In fact, is it even possible to
dispense with ordinary language? If these two questions be
answered, ‘No, they cannot,’ and ‘Yes, it is possible,’ there remains
the final question, Does Russell’s artificial language succeed?
Some of what follows may suggest that occasionally Russell’s
language is worse than the one he discards.
Let us then state some of the specific difficulties, beginning
with one of the standard medieval insolubilia that Russell men-
tions, viz., the case of the Cretan liar. The problem assumes that
Cretans never speak the truth. Everything they say is a lie. We
now meet a Cretan who tells us “I am a liar.” But for this to be
true, as it is assumed to be, it must be false. Worse, if this be false,
as it is not, it is certainly true, for if the liar is lying, he has
certainly told the truth.
Ordinary liars, of course, sometimes speak the truth; and if one
such says, “I am a liar,” he is telling the plain unparadoxical
truth. But Cretans, so we assumed, tell nothing but lies. Before
Russell’s solution is given, and before other difficulties are enu-
merated, ordinary language might reply: On the assumption that
BERTRAND RUSSELL 13

Cretans tell nothing but lies, it is impossible that any Cretan

should say, ‘Iam a liar.’ The so-called paradox arises because the
paradoxer asserts two contradictory propositions. It is impossible
for both to be true. Therefore, the person who wishes to puzzle us,
if he is to say anything meaningful, must choose between asserting
that all Cretans tell nothing but lies and the supposition that one
Cretan says, ‘Iam aliar.’ The laws of logic, particularly the law of
contradiction, forbid the insolubilist from making both state-
ments. And if he does not make both statements, there is no
difficulty. To put it simply: No such Cretan would admit that he
was a liar.
There is also the clever puzzle about the barber who shaves
only those, but all of those, who do not shave themselves. Does this
barber shave himself or not? Well, of course, if he shaves himself,
he cannot shave himself, for he shaves only those who do not shave
themselves. But if he does not shave himself, he must shave
himself, for he shaves all those who do not shave themselves.
Russell admits that this puzzle is not too hard to solve; but to the
present writer there seems to be a different and easier solution
than the one Russell offers. As in the case of the Cretan liar, this,
too, is a disguised contradiction. Its suppositions or requirements
are logically incompatible. Hence, there is no paradox. The person
propounding the puzzle is simply making two statements that
cannot both be true. Hence, there is no need of an ideal, artificial
language with complicated formulas. But Russell wants to asso-
ciate it with other forms that resemble Plato’s ‘Third Man’
argument.
Before this ancient and more complicated problem is taken up,
several more preparatory considerations can contribute to the
background material. Each one makes some advance to more
technical levels. This one concerns Russell’s ‘pure form of all
general propositions.’ A completely general proposition is one
which contains only variables. Russell gives a series of successive
generalizations:
Socrates loves Plato
x loves Plato
x loves y
xeRy
14 BERTRAND RUSSELL

Asa can mean any manor any thing, so R can mean any relation.
This final form is so general that it is the pure form of all general
propositions. Then Russell continues, “Suppose I say: ‘v R y
implies that x belongs to the domainof R. . . .. You might think it
contains such words as ‘belong’ and ‘domain’ [and we might also
add ‘imply’], but that is an error. It is only the habit of using
ordinary language that makes these words appear. They are not
really there.”> One might be excused for thinking that they are
really there; but other matters call for consideration before criti-
cism begins.
Another example of confusions due to ordinary language comes
in Lecture VI, Descriptions and Incomplete Symbols. Russell
wishes to distinguish between a name and a description. There
are several instances mentioned; e.g., “Romulus is not really a
name, but a sort of truncated description.” More space, however,
is given to ‘Scott,’ which apparently is at least sometimes a name,
and ‘the author of Waverly,’ which is obviously a description. The
reason that this phrase is not a name is that anyone who under-
stands the English words separately immediately understands
the phrase also. But the word ‘Scott’ gives us no information atall.
That is, the four words ‘the author of Waverly’ have had their
respective meanings fixed in the language before ever they were
put together in this phrase; but let all other English words be so
fixed, yet they provide no meaning for ‘Scott.’
Now, there is nothing surprising, nor particularly useful, in
distinguishing names from phrases, unless Russell continues by
analyzing the statement ‘Scott was the author of Waverly.’ In one
place his analysis is: there is an entity c such that the statement ‘x
wrote Waverly’ is true if x isc and false otherwise; and ¢ is Scott.
To the uninitiated this may sound doubtful, and certainly sounds
awkward. Surely it is no improvement over ordinary English.
But what is not so awkward and by no means trivial is his conclu-
sion that this clears up two millennia of muddle-headedness about
‘existence, beginning with Plato’s Theaetetus.
How the concept of existence comes to intrude here requires
further remarks on Sir Walter. Also involved is the meaning of

>The Philosophy of Logical Atomism, ed. Marsh, pp. 238-239.

BERTRAND RUSSELL 15

the word is. In Logical Atomism (1924, ed. by A. J. Ayer, Logical

Positivism, p. 38) he complains that Western languages are con-
structed on a subject-predicate form. Non-Aryan languages do
not need subjects and predicates, except in connection with
Buddhist theology.® This accounts for subject-predicate logic and
substance-attribute metaphysics in Greek.
Before returning to Scott one might note that poor old Socrates
had a most difficult time explaining to his fellow Greeks the
difference between a universal and an individual. Even after
Hippias admits that justice, wisdom, and goodness are “some-
thing,” and that beauty is “something real,” he replies to the
question, “What is beauty—not what is a beautifulthing?” @u ri
€oT. KaAGY, ‘aA’ 6 TL €or TO KaAOr), “a beautiful girl” (Greater
Hippias, 287c-e). Could it not be that human rationality led man
to use subjects and predicates, rather than subjects and predicates
having deceived us regarding universals?’
Sir Walter Scott, therefore, plus existence and the analysis of
propositions must continue to be considered. Russell’s proofs that
the phrase is not a name are as follows: “In ‘Scott is the author of
Waverly’ the ‘is,’ of course, expresses identity, i.e. the entity whose
name is Scott is identical with the author of Waverly. But, when I
say ‘Scott is mortal’ this ‘is’ is the ‘is’ of predication, which is quite
different from the ‘is’ of identity.”®
The next of these several points, all of which are related in one
way or another, is Russell’s contrast between verbs and the copula.
Traditional logic reduces sentences in which ordinary verbs
occur, such as, ‘Men think,’ to ‘Men are thinkers.’ ‘Men’ is the
subject, and ‘thinkers’ is the predicate. Hence all arguments can
be symbolized and tested for validity by putting them in some
such form as ‘A(ba) A(cb) implies A(ca).’ For Russell the subject-
copula-predicate scheme has, if not validly, at least psychologically,
os

6A friend of mine who has lived with the Navajos, speaks their language, and
has translated English books into Navajo, reports that he can find no basic
difference between the structure of Navajo thought and the logic of Greek or
English. ;
7Russell indeed addresses himselfto this question in the immediately following
pages; but he seems to have altered it somewhat so that his answer does not
apply.
8The Philosophy of Logical Atomism, ed. by Marsh, pp. 244ff.
16 BERTRAND RUSSELL

led philosophers into the confusions of substantialism. Now, if

this is merely a psychological mistake, it is irrelevant. Everyone
makes mistakes. Russell really ought to insist that the subject-
copula-predicate scheme forces Aristotelian metaphysics on
everyone who uses it. This is hardly credible; but at any rate the
larger problem focuses our attention on the question whether
propositions always have predicates, or whether sometimes
ordinary verbs cannot be so reduced.
One of Russell’s examples where the copula-predicate analysis
cannot fit is the relation ‘is greater than’; for example, ‘three is
greater than two.’ This example is cited in a paragraph opposing
Hegel’s all-embracing, internal relation theory. Symmetrical
dyadic relations can indeed be reduced to sameness of predicate;
but with asymmetrical relations it is impossible. This impossibil-
ity, says Russell, “is a matter of a good deal of importance...
because a great deal of traditional philosophy depends upon the
assumption that every proposition really is of the subject-
predicate form, and that is certainly not the case” (Philosophy of
Logical Atomism, p. 207).
To dull the force of Russell’s contention, an example of an
asymmetrical relation reduced to the subject-copula-predicate
scheme and put in syllogistic form may help. Take the inference:
three is greater than two, two is greater than one, therefore three
is greater than one. The ‘is’ here is not the copula, but the phrase
‘is greater than.’ The inference is not syllogistic because if it be
put in copula form, namely,
(Three) is (greater than two)
(Two) is (greater than one)
there is no middle term. However, this does not prove the impos-
sibility of putting the argument in syllogistic form; namely,
All three’s are greater than two’s;
All greater than two’s are greater than one’s;
Therefore, all three’s are greater than one’s.
As was said before, one of the reasons Russell wants verbs
instead of copulas is that the copula form leads to the Aristotelian
concept of substance. But it is not at all clear that such is the case.
Aristotle himself allows quantities, qualities, relations and the
other categories to serve as subject concepts in propositions; and
BERTRAND RUSSELL WH

this does not make a quality a substance. In one of Aristotle’s

syllogisms the twinkling of a star is a subject, and twinkling no
doubt falls under the category of action. Therefore, to say nothing
stronger, Russell’s argument against traditional logic on this
basis is unsuccessful.
Involved in all this is Plato’s ‘Third Man’ argument. The
Parmenides expounds an objection to the theory of Ideas: if the
similarity between Socrates and Crito requires us to posit the
Idea, Man, then the similarity between Socrates-Crito and Man
requires an Idea-prime, or Third Man. Since this initiates an
infinite regress, the Ideal theory is untenable. Plato leaves all but
one of seven objections unanswered; and this is not that one. But
does it not seem strange that a genius of Plato’s stature would
have left them unanswered unless he had believed the answers to
be readily discoverable? Especially since he continues with the
Ideal theory in his subsequent dialogues. In the Parmenides he
signs off with “a very brilliant man will be able to understand
that there is a genus for each thing and an absolute reality per
se.... But if anyone denies the existence of Ideas of things,
because of the objections above and similar ones [!], . . . he will
not know how to conduct his thought . . . and thus he completely
destroys the possibility of argumentation.”
This of course does not prohibit Russell from making the
attempt. “You can start,” Russell says (Marsh edition, p. 259),
“with the question whether or not there is a greatest cardinal
number.” To say no, results in the curiosity that there are more
numbers than there are [other] things in the world. Infinity may
indeed be a curiosity for some people; but a little arithmetic
shows that it is always possible to add one, and hence the number
series is infinite. Of course, if the other (Russell omits this word)
things happen to be infinite, instead of finite as Russell seems
silently to suppose, the cardinal numbers and the other things
would be equal, so that one of them would not be greater than the
other. Without considering this minor flaw, for it only applies to
the uneducated curiosity seeker, Russell argues that particulars
(individual things?) and classes do not exist in the same sense.
The reason is that a world of three particulars would produce
eight classes (ab, ac, cb, abc, as well as a, b, c, alone and zero), and
therefore this world would have eleven things (existents?). But
18 BERTRAND RUSSELL

when he concludes, “That, on the face of it, seems to land you ina
contradiction” (p. 260), we can admire his quip, “There are fewer
things in heaven and earth than are dreamt of in our philosophy,”
but we cannot accept his logic: it just does not follow. It would not
follow, so it seems to one person, even were the ambiguities which
are indicated in the parentheses above removed.
If, however, Russell’s argument still seems valid or at least
plausible to others, the point may be better determined by an
analysis of Russell’s further explanation. He asks us to consider
those classes which are not members of themselves. “You would
say generally that you would not expect a class to be a member of
itself. For instance, if you take the class of all the teaspoons [Man]
in the world, that is not itself a teaspoon [a man].’° In ancient
language this means that a sensory individual, if there are any
such, is not an Idea. But Russell seems to shift unwittingly toa
different problem. Instead of continuing with sensory individuals,
his discussion concerns the relationships among classes. As with
the barber, he now asks, Is the class of classes that are not
members of themselves, a member of itself or not? “Suppose that
it isa member of itself. In that case it is one of those classes that
are not members of themselves, i.e., it is not a member of itself.
Let us then suppose that it is not a member of itself. In that case it
is not one of those classes that are not members of themselves, i.e.,
it is one of those classes that are members of themselves, i.e., itisa
member of itself” etc. (p. 261).
The answer to this, so it seems to the present writer, is that
every class is a member of itself. Were this not so, logic would be
impossible. In fact, Russell himself says so. The symbolic logic he
desires to substitute for ordinary language depends on the axiom,
a <a. All the a’s are a’s. One is included in one, and zero is
included in zero. Anything else, as Parmenides said, “completely
destroys the possibility of argumentation.”
There is one final remark, or two, concerning an artificial ideal
language, which might be postponed until the completion of the
material on Wittgenstein, for it applies to him as well, but which
can be placed here because it attaches to a sentence in Russell’s

°The Philosophy of Logical Atomism, ed. by Marsh, p. 260.

BERTRAND RUSSELL 19

Preface to the Tractatus (tr. by Pears and McGuiness, p. x).

Russell states, “In speaking of a ‘complex’ we are, as will appear
later, sinning against the rules of philosophical grammar, but
this is unavoidable at the outset.”
Here Russell admits that ordinary language is indispensable.
No artificial language can be constructed without it. But in
addition—what Russell did not say—artificial language is not
indispensable. Not strictly, formally, and rigorously indispensable.
Symbolic logic, like arithmetic and algebra, is a useful tech-
nique. Modern civilization could not have developed without
them. But this does not mean that ordinary language can be
discarded. On the contrary, ordinary language not only is indis-
pensable for the construction of symbolisms; symbolisms can
always be translated back into ordinary language. To take an
easy example: from the equation a2 + 2ab + b? = Othe valueofais
neatly determined by the rules of quadratics. We would not want
to do without them; and it is possible that the solution would never
have been discovered without the symbolism. But it could have
been; and the equation can be put into ordinary, if awkward,
English. The English would go: a number multiplied by itself
added to double its product with another number and then added
to that other number multiplied by itself equals zero. Now, noone
wants to talk this way. It is much easier to write a short line of
symbols instead of two or three lines of English. But the philo-
sophic point is that not only can the equation be expressed in
English, but that without ordinary English the equation could
never have been understood. Plus, exponent, multiplication,
equality had to begin in ordinary language. And even today a
small amount of English appears here and there to indicate what
some part of a formula means.
The arguments now completed seem sufficient to dispose of the
thesis that ordinary language should be abandoned and replaced
by an artificial symbolism. Two things, however, remain to be
done. First, there are metaphysical, as well as linguistic, assump-
tions that underlie this desire for an ideal language. And second,
there is the symbolic logic itself, the perfection of which requires
evaluation.
Now, first, the brief intellectual biography at the beginning
noted that Russell early renounced Hegelianism and became an
20 BERTRAND RUSSELL

empiricist. This change started with an attack on Bradley’s, and

Hegel’s, theory of internal relations and the substitution of an
atomic theory of external relations. The former, holding that
everything is implicated in everything, results in an absolute
monism. The definition of cat, for example, is part of the definition
of dog, and also of Betelgeuse. For Russell relations are external to
the objects related. These relations, though it seems strange to
say so, are grasped by immediate sense perception. This seems
strange because it is hard to see what color above and to the left of
are, or to hear what noises wncle and is greater than give off.
However, such are the atoms of Russell’s world.
In conformity with this, propositions are true in isolation. A
proposition is true if it corresponds to an atomic fact or a combi-
nation of them. ‘The car is in the garage’ is true if we see a car, a
garage, and an in. Thus, language consists of words each of which
designates a sensory individual.
To be fair to Russell, one must acknowledge that he later
modified such an absurdity. He came to doubt the reality of 7s and
the, if not in. These nonrealities he then explained as the logical
positivists did later, as parts of a logical framework without
objective referents. This framework became his symbolic logic.
An earlier section of this chapter noted that Russell changed his
mind every so often. The critic cannot simply say, This is Russell’s
view. In fact, he even changed his meaning of the word this. The
critic therefore, at least this critic, can hope only for reasonable
accuracy in repeating some of Russell’s views and then analyzing
those chosen. These analyses are supposed to help in any con-
structive work the reader is inclined to attempt.
It does seem, however, that over a long period Russell believed
in the logical independence of every fact and the theory of external
relations. It is hard to think of any empirical philosophy that can
believe otherwise. Nevertheless Aristotle the empiricist, not to
mention Hegel or Bradley, notes phenomena that ill accord with
logical atomism. Uneducated people talk about the five senses,
and touch is one of the five. But Aristotle knew that what we call
touch is three different senses. He explained the common mis-
apprehension on the ground that the skin is not the sense organ,
but a medium that serves three different organs underneath.
Now, if the air, continues Aristotle, were a part of the body,
BERTRAND RUSSELL 21

enveloping the face as the skin does the fingers, we would suppose
that smell, taste, hearing, and sight are all one. Even as it is,
though Aristotle does not mention it, we cannot be sure that sight
is asingle sense. Maybe there are as many senses as there are rods
and cones in the retina. The difficulty here is in identifying an
atomic sense. Russell himself lamely replies, this is as simple as I
can now make it.
The troubles with atomism, however, are still greater. Is any
proposition true in isolation? Would an atom by itself be the same
regardless of how the rest of the world might change? There are
plausible examples that it would not. Here is a rock that weighs
six pounds. But if an astronaut carries it into space it weighs
approximately zero. When he drops it on the moon, it weighs one
pound. The truth of these propositions depends on the relation of
the rock to the other parts of the universe. No one is true in
isolation. Obesity is cured by a trip to the moon.
Another example is a piece of canvas painted half red and half
green—or any other two colors. Through these two halves of the
canvas paint a stroke of gray, a mixture of black and white; but it
will not be gray on the canvas. The single stroke of paint will be
one color on the top half of the canvas and a different color on the
bottom half. Since everything seen has a background, its colorisa
function of its background. It is false to say it remains what it is
no matter how the rest of the universe changes.
One further example. If there were no sense of sight, there
would be no sense of hearing. If there were nothing hard, there
would be nothing soft. If there were no animals, there could be no
plants. The reason is that each of these terms expresses a distinc-
tion from its opposites. Sight is a form of nonhearing. Were they
the same, we might have the term sensation, but we would not
have two terms of different meaning. The terms ‘plant’ and
‘animal’ would not apply to different objects, if there were no
different objects. There might be ‘living beings,’ but no plants
and animals. Similarly, there would be no living beings, if there
were no nonliving beings. This should be sufficient to dispose of
logical atomism.
The final section on Russell must now dwell on the basic propo-
sitions of his artificial language—the first few steps in his
symbolic logic. George Boole, the inventor of symbolic logic, may
22 BERTRAND RUSSELL

have aimed to express Aristotelian logic in symbols. Can the

square of opposition, its contradiction, contrariety, subalterna-
tion, and subcontrariety be preserved, plus, of course, obversion,
simple conversion, and any other elementary terms? What about
categorical forms? What does it mean to say that ‘All Athenians
are Greeks’? What does all mean? But between George Boole’s
original attempt and Russell’s Principia Mathematica, the logi-
cians had concluded that this could not be done.
Russell explained that ‘all men are mortal’ means “f anything
is a man, it is mortal.’ It does not mean that there ave any men.
However, the phrase ‘Some men are mortal’ means ‘there exists at
least one man and he is mortal.’ Hence if all men are mortal, it
does not follow that some are. If all dogs are canines, it is invalid to
infer that some dogs, for example, bull dogs, are. What has
happened is that the symbolism invented between 1850 and 1900,
although it preserved contradiction and obversion, made sub-
alternation a fallacy.
A little symbolism, just a little, explains how. rirst the terms
‘zero’ and ‘one’ were introduced. If any two classes may be multi-
plied together, as they must to preserve generality, that is, joined
together by the conjunction ‘and,’ such as x is both a dog and
brown, it follows that the multiplication of contradictory or
contrary classes results in the zero class, the null class, a class
said to have no members, including nothing. Then next, since two
contradictories exhaust the field, the addition or disjunction of
two contradictories results in the number one, the universe, said
to include everything. For that matter, since every class must
have acontradictory, to preserve generality, the contradictory of
zero is one. Zero contains nothing; the universe, or the universe of
discourse, contains everything: it is the class that exhausts
whatever the subject of discussion happens to be. With this the
proof that subalternation is a fallacy is as follows.

A(ab)=a<b (definition)
This reads, class a is included in the class 6. Therefore, by
obversion

E(ab)=a< 0’.
The sign <, when used in mathematics, means ‘is less than’; and
BERTRAND RUSSELL 23

this is a factor in developing mathematics from symbolic logic.

Similarly, the plus sign of mathematics indicates disjunction in
logic because

(a + b)c = ac + be,
that is, cat and either auburn or black equals an auburn cat ora
black cat. The ac and be are examples of multiplication: the object
is both a cat and a black object.
The refutation of subalteration now proceeds.
Allaisb implies Some a is}
A(ab) < I(ab) (substitution)
(a6) <= (a <6)’.
Since E(ab) is (a < 0’), its contradictory J is (a < b’)’. Since
further a and b are variables and can take on any meaning, the
next line is a special case of the previous line.
io < oy) < (0 Soy.
But the contradictory of zero is one; therefore
(0o<o)<(0<1y.
Since every class is included in itself, (0 <0) must be a true
statement. But since the universe includes all classes, zero must
be included in the universe. Its negation is false. Hence the last
line of the symbolism has a true premise and a false conclusion.
Therefore all does not imply some.
From these definitions of the categorical forms it is possible to
develop an extensive symbolic logic. But instead of providing us
with twenty-four valid syllogisms, it gives us only nineteen. There
is no fallacy in this development. Everything follows rigorously
from the initial definitions. But it is a restricted system. It is like
a geometry that has only seven lines with three points on each
line. Theorems can be deduced, but too few.
In other words, modern logic has failed to put Aristotelian logic
into symbolic form. Its language cannot say as much as ordinary
English can.
The source of the flaw is the initial definition. If (all a is 6)
means (ais included in b), subordination cannot be defended. But
24 BERTRAND RUSSELL

should not this result have prompted the logicians to find a

formula for all that would have expressed the English meaning?
After all, (a < b) is an arbitrary choice. It is Russell’s choice.
Anyone else is as free as Russell was to choose a different defini-
tion. By adifferent definition all can imply some. Such a definition
might be
(a<b)[(b<a)+(a<b’) (b' <a’).
Corresponding to Euler’s diagrams rather than to Venn’s, this
formula will produce
AE <o therefore

A-< i.
There is some difficulty with this formula. It would require us
to treat zero much as it is treated in arithmetical division.
Whether this is fatal or not, one must decide for himself. However,
there isa more pointed difficulty with Russell’s definition than its
arbitrary nature. When Aristotle said, All the a’s are b’s, he
meant that every a isa b. But when Russell asserts, as the defini-
tion of all, that (o < 7), he cannot say that every zero isaone. This
means that Russell’s definition of all does not reproduce the
English meaning of all. If then Russell is not talking English,
perhaps when all dogs are canines, some dogs may not be
canines.
The conclusion is that artificial languages have certain degrees
of utility. But ordinary language is indispensable. Perhaps it
should be added that a person should say what he means, or at
least mean what he says.
 7A

Ludwig Wittgenstein (1889-1951)

ALTHOUGH it was impossible to cover all of Russell’s philosophy,
for it is extremely extensive, someone might with difficulty do
justice to Wittgenstein. The difference lies in the fact that while
Russell changed his mind with every succeeding volume, Wittgen-
stein made but one major shift, repudiating the early Tractatus
Logico-Philosophicus and substituting his Philosophical Investt-
gations. Born in Vienna, he came to Manchester in 1911, then
went to Cambridge where he studied under Russell, and though
serving in the Austrian army, published his Tractatus during the
war. He returned to England in 1929. During 1932-19384 he
wrote his Philosophical Grammar, in which he seems to make a
step or two toward his later views. Wittgenstein died in 1951 and
his Philosophical Investigations were published posthumously in
1953, a partly dictated and partly corrected edition of his students’
lecture notes.
As Russell was never a logical positivist, so Wittgenstein was
not a strictly orthodox member of the Vienna Circle. Both men
indulged, perhaps more than they realized, in metaphysics; and
Wittgenstein added a touch of mysticism ill in accord with the
ideals of unified science. From the Tractatus: “6.44 It is not how
things are in the world that is mystical, but that it exists.” The
conflicting tendencies toward logical positivism and toward
Platonic mysticism, if this phrase is at all permitted, are found in
My Mental Development (in Schilpp, ed., p. 12). Wittgenstein
writes, “Bradley argued everything common sense believes in is
mere appearance; we reverted to the opposite extreme and
thought that everything [ital. his] is real that common sense,
uninfluenced by philosophy or theology, supposes real. ... We
25
26 LUDWIG WITTGENSTEIN

allowed ourselves to think that grass is given... and also that

there is a pluralistic timeless world of Platonic ideas.” Does this
mean that a timeless world of Platonic ideas is the view of common
sense, uninfluenced by philosophy or theology? But let it pass.
With such caveats Wittgenstein is closely related to the Vienna
Circle. Apparently it was from him that they adopted the tauto-
logical view of logic and mathematics. They both held that the
role of philosophy is precisely the clarification of language.
Wittgenstein does indeed have a tinge of metaphysics, and among
other things accepts Russell’s theory of external relations (4.122);
nevertheless toward the end of the Tractatus he is frequently
critical of Russell (3.331, 3.333, 5.535, 5.5422, 6.123, 6.1232) and
had earlier enunciated very positivistic theses, namely, “4.0031
All philosophy is a critique of language.” “4.111 Philosophy is not
one of the natural sciences. ... 4.112 Philosophy aims at the
logical clarification of thoughts. Philosophy is not a body of
doctrine but an activity. .. . Philosophy does not result in ‘philo-
sophical propositions,’ but rather in the clarification of proposi-
tions. .. . 4.113 Philosophy settles controversies about the limits
of natural science.”
These theses and others like them not only limit philosophy toa
study of language, but also limit knowledge to the results of the
positive sciences. Other so-called philosophic or religious lan-
guage is nonsense (cf. 4.008).
These introductory remarks have brought us partway into the
Tractatus. Perhaps it is best to return to its beginning. The
opening propositions of that great work hardly encourage a
reader to proceed. The machinery clanks. “1. The world is all that
is the case. 1.1 The world is the totality of facts, not of things.”
Presumably this defines the world by equating “the case” with
“facts,” and distinguishing them from “things.” “1.12 For the
totality of facts determines what is the case.” These “facts” exist in
“logical space.” Logical space is not further explained.
Aside from the fact (?) that “fact” and “case” are identical,
making one of the two terms superfluous, the same identity makes
it strange to say that “the totality of facts determines what is the
case.” There is really no determination: the words fact and case
are synonymous.
LUDWIG WITTGENSTEIN rea

Proposition 1.21 gives something more substantive: “Each item

can be the case or not while everything else remains the same.”
This is a theme common to Russell, Wittgenstein, and the logical
positivists. But it is neither an empirical discovery nora tautology.
It isa universal proposition, as universal as any proposition could
possibly be, for it asserts something about every item in the
universe, past, present, and future. As such it is a metaphysical
principle accepted as a substitute for Hegel’s internal relations.
Now, Russell openly admits some metaphysics; Wittgenstein does
so less openly; but the strict logical positivists, as will be plain and
evident in the next section on Carnap, abhor all metaphysics with
a holy hatred—no, unholy, for holy is too theological. But the
theory of a pluralistic universe is itself metaphysical. As a uni-
versal proposition it cannot be established by finite observations.
As metaphysical it should be ruled out by the positivistic theory
of language. But it permeates logical positivism.
In case a case, a fact, a thing, and the items of the preceding
quotation are unclear, Wittgenstein continues: “2. What is the
case—a fact—is the existence of states of affairs. 2.01 A state of
affairs (a state of things) is a combination of objects (things).”
But if a state of affairs is a state of things, identifying affairs
and things, and if afact isa state of affairs, how can the world bea
“totality of facts, not of things”? This confusion would lead the
most recalcitrant to favor an artificial, ideal language, or else a
better command of German.
Wittgenstein’s metaphysical foundation comes to the surface
every so often. Proposition 2.0123 is as Hegelian as Hegel himself
could wish. It reads, “If I know an object [a thing] I also know all
its possible occurrences in states of affairs [states of things].
Every one of these possibilities must be part of the nature of the
object. A new possibility cannot be discovered later.” Indeed, this
is mere Hegelian than Hegel himself. The great absolute idealist
held that the simplest object implicitly contains the universe. If it
were different in any way, every other thing in the universe
would be different. It would be a completely different universe.
But Hegel never asserted that knowledge of a single thing is itself
an explicit knowledge of “all its possible occurrences.” There
must be an analytic or dialectic process to bring these implicit
relationships to light.
28 LUDWIG WITTGENSTEIN

Wittgenstein offers further metaphysics: “2.024 Substance is

what subsists independently of what is the case. 2.025 It is form
and content. 2.0251 Space, time, and color (being colored) are
forms of objects. . . . 2.027 Objects, the unalterable, and the sub-
sistent are one and the same. 2.0271 Objects are what is unalter-
able and subsistent; their configuration is what is changing and
unstable. 2.0272 The configuration of objects produces states of
affairs. . . .2.062 From the existence or non-existence of one state
of affairs it is impossible to infer the existence or non-existence of
another.”
At the least this section is puzzling.! Whatever is symbolized by
the new term substance is independent of the “case,” the “facts,” a
“state of affairs,” or “state of things.” Yet “substance” is “form and
content.” Now, if all reality consists solely of visible “things,”
‘being colored’ may be an unalterable “form” of objects; but surely
“context” includes the particular color at a given time; and this is
not unalterable. In proposition 2, Wittgenstein seemed to identify
things, affairs, and objects. Here he says they are unalterable.
But what can a thing or state of affairs be, if it is unalterable?
This leads to a most important matter underlying not only
secular positivism, but also many of the later theories of religious
language. The aim here will be to refute a fundamental flaw in
empiricism, and the argument should be kept in mind all the way
through to the end.
The standard forms of empiricism, surely Wittgenstein’s,
depend ona theory of images; and they usually add an Aristotelian
process of abstraction in order to get concepts. The following
texts support this assertion:
2.1 We picture facts to ourselves. . . . 2.13 A picture is a model of
reality. 2.13 In a picture objects have the elements of the picture
corresponding to them. 2.131 In a picture the elements of the
picture are representatives of objects... . 4.01 A proposition is a
picture of reality ...a model of reality as we imagine it.
4.011 [printed music does not seem to be a picture of the reality, the
sound] and yet these sign languages prove to be pictures, even in

'Max Black, in A Companion to Wittgenstein’s “Tractatus” (Ithaca, New York:

Cornell University Press, 1964) remarks that 2.062 seems to conflict with 2.05.
LUDWIG WITTGENSTEIN 29

the ordinary sense, of what they represent. 4.012... aRbstrikes

us as a picture. In this case the sign is obviously [!] a likeness of
what is signified. .. . 4.016 [hieroglyphics depict the facts they
describe] and alphabetic script developed out of it without losing
what was essential to depiction. . . .4.06 A proposition can be true
or false only in virtue of being a picture of reality.

Since this pictorial view is inherent in all empiricism, excepting

only Berkeleyan subjective idealism and pyrrhonian skepticism,
it deserves the most careful consideration. Important as the theme
is in philosophy, there is nothing more important for religious
theories of language and knowledge, for it underlies the possibil-
ity of any and every theological sentence.
Other minor criticisms may surface on later pages, but here
two fundamental objections demand attention: one has to do with
the idea of picture, and asecond with the idea of representation or
correspondence. This second point, the ‘correspondence theory of
knowledge,’ faces the insuperable objection that it disallows any
knowledge of reality at all. Whatever reality may be, whether
individuals like trees and rocks, or Platonic Ideas, or whatever,
this theory provides us only with pictures of them. The object of
knowledge is therefore a representation and not the reality itself.
Since the mind contains only the picture and never the ‘thing,’
there is no possibility of knowing whether the representation is
similar to the object or not. To recognize a similarity between two
things, they must be compared, and hence both must be in the
mind. But if the reality is in the mind, the picture with its
similarity is useless. If the reality is not in the mind, the picture,
so far as we know, is a picture of nothing. There is hardly any
objection to empiricism more fundamental than this one. But
there is another, not much less important, whose force seems to be
less generally recognized.
Further quotations: “2.0141 A picture is a fact. . . . 2.151 Pic-
torial form is the possibility that things are related to one another
in the same way as the elements of the picture. 2.1511 Thatis how
a picture is attached to reality; it reaches right out to it.”
After a number of other propositions in which pictures seem to
include logical forms as well as spatial things, proposition 3.1431
reads, “The essence of a propositional sign is very clearly seen
[ital. mine] if we imagine one composed of spatial objects (such as
30 LUDWIG WITTGENSTEIN

tables, chairs, and books) instead of written signs. Then the spatial
arrangements of these things will express the sense of the
proposition.”
In these numbered propositions the idea of comparison is men-
tioned, but in no such way as to avoid the argument against all
correspondence theories of knowledge. The other point, pre-
viously mentioned, has to do with the idea of images and their
special arrangements. The importance of this may be emphasized
by returning to something not previously quoted from Bertrand
Russell. In Logic and Knowledge (ed. by R. C. Marsh, p. 293),
discussing Watson and behaviorism, Russell wrote, “There is a
valid objection to the behavioristic view of language on the basis
of the fact... [viz.,] the denial of images appears empirically
indefensible. ... If you try to persuade an ordinary uneducated
person that she can not call up a visual picture of a friend sitting
inachair... she will conclude that you are mad. (This statement
is based upon experiment.)” Then he adds, “To ‘think’ of the
meaning of a word is to call up images of what it means” (p. 300).
To emphasize further the importance of this, we quote the
source in Hume—though Thomas Aquinas and Aristotle said
much the same thing. Hume put it this way: “That idea of red,
which we form in the dark, and that impression which strikes our
eye in sunshine, differ only in degree, notin nature. That the case
is the same with all our simple impressions . . . everyone may
satisfy himself... by running over as many as he pleases” ( Trea-
tise of Human Nature, I, i, 1).
There are two objections to this Hume-Russell-Wittgenstein
position. First, it is hard to understand Russell’s assertion, “This
statement is based on experiment.” Neither Hume nor Russell
could examine the “everyone” Hume mentions. Was Russell’s
experimentation limited to one woman sitting ina chair? At best
it is an induction from questions asked of and answers received
from a large sample of “ordinary uneducated persons.” This is an
induction; and the validity of induction is an indispensable
element in scientific positivism. But induction, unless it be
complete induction which is never the case in science, is always
invalid. Many people have seen a hundred black crows without
ever having seen an albino. Their induction that ‘All crows are
black’ is a mistake. Less persuasive but equally applicable, is the
LUDWIG WITTGENSTEIN 31

induction by which a physicist formulates the law of the lever.

One might believe that all levers ‘obey’ the equation, but the
induction does not guarantee it. As a matter of fact, whenever a
physicist formulates a law, he uses a number of a priori, unsup-
ported assumptions. Furthermore, this “whenever” is not based
on an incomplete induction, but upon an analysis of laboratory
methodology.
But second, and conclusively, induction, if it proves anything,
proves the falsity of the empirical principle. Brand Blanshard
reports one carefully made induction.2 The present writer also,
though not with mathematical accuracy, has obtained some per-
centages by questioning students over a period of twenty years or
more. Ina course on modern philosophy, when the class got to the
chapter on Hume, he would ask, “Do you imagine red in the dark
as Hume says you must?” and “Can you see in your mind an absent
friend sitting in the chair before you?”
Over twenty years the number of those who answered ‘no,’ to
those questions is about five percent. This is the albino crow that
wrecks empirical ornithology. Further, among the ninety percent
with visual imagery an interesting different was reported. Some
of them dreamed in technicolor, others only in black and white.
Here, then, is imagery, but not the sort of imagery that Hume
insisted everybody has.
Now, someone might suspect that a mistake had been made in
every case within the five percent. Hardly likely; for this would
give one hundred inexplicable mistakes. What further supports
the conclusion that several people have no visual imagery are the
answers to similar questions about auditory imagery. The exper-
iments showed that ten to fifteen percent had none. Many more,
at least fifty percent, had no gustatory or olfactory imagery.
Tactual imagery came between auditory and olfactory. This is
most convincing to the present writer, for he, too, cannot find any
irfiage of red almost as vivid as the red seen in sunlight. To put it
accurately, he can find no image of red at all; and to suppose that
he can ‘see’ an absent friend is ludicrous.

2Brand Blanshard, The Nature of Thought, two vols. (London: George Allen and
Unwin Ltd.), vol. I, pp. 260-263.
By LUDWIG WITTGENSTEIN

This unimaginative position does not solve all the problems of

language; but it will be the basis for rejecting every empirical
theory of knowledge and will play a major role in defending the
legitimacy of religious language. It entails a different meta-
physics; a different view of what reality consists of; a different
view of how one reality is related to another. But first there is
more to consider in the positivistic systems.
What now follows will be inasense desultory and unsystematic.
The aim is to show how or that logical symbolism, which most
people find exceedingly tedious, affects one’s views on ethics and
theology.
The Tractatus gives an early limitation on language: “3.221
Objects can only be named. Signs are their representatives. I can
only speak about them: I cannot put them into words. Propositions
can only say how things are, not what they are.” Emil Brunner
was later to say, If you talk about God, you are not talking about
God.
When Wittgenstein also says, “4.001 The totality of propositions
is language,” he may be defining language, but at any rate he is
asserting that Robert Burns did not make much use of language.
Many times Wittgenstein’s remarks are the quintessence of
wisdom; for example, “4.002 .. . The tacit conventions on which
the understanding of everyday language depends are enormously
complicated.” Indeed, they are so complicated that one begins to
hope for a perfect artificial language. In a logic class one day the
instructor defined aterm, using five key words. The first student
who spoke showed by what he said that he had missed the signifi-
cance of one of those five words. The word was further explained.
Then in turn four other students showed how little they had
understood each of the other four words. And this was ina college
classroom, where they expected or should have expected a certain
degree of preciseness. Imagine then how sloppy, how confused,
how unintelligible ordinary conversation is.
But though this is a lesson all college students, and many
college professors, should learn, did Wittgenstein mean no more?
His previous lines are, “Language disguises thought. So much so,
that from the outward form of the clothing it is impossible to infer
the form of thought beneath it because the outward form of the
LUDWIG WITTGENSTEIN ao

clothing” in the nineteenth century was not the bikini style of the
late twentieth. But if it is impossible, not just usually but always,
how does one ever, even once, understand what another says, or
where can one begin his search for logical form? Perhaps
Wittgenstein’s language here obscured his thought, for later he
says, “4.116 Everything that can be thought at all can be thought
clearly. Everything that can be put into words can be put clearly.”
Wittgenstein does indeed limit what can be thought and what can
be put into words, but at least, so it seems, some things can be
thought and can be said clearly.
A puzzle arises when Wittgenstein acknowledges that some
things can be “shown” but cannot be expressed (see 4.12-4.126).
At that point the matter is quite technical; but later he extends
the application: “5.62... What the solipsist means is quite
correct; only it cannot be said, but makes itself manifest.” Here
then is a meaning, and therefore a thought—for can a solipsist
mean something without thinking it?—yet it is a thought that
cannot be put in words. Admittedly in 4.116 Wittgenstein did not
say, ‘What can be thought can be said.’ Thus he escapes the
obvious contradiction. But the puzzle remains that he has aclear
thought that cannot be said. This is hard to think.
Wittgenstein follows this puzzle into the spheres of ethics and
aesthetics, and into theology as well. Consider the following
lengthy quotation with care.
6.41 The sense of the world must lie outside the world. In the world
everything is as it is, and everything happens as it does happen; in
it no value exists—and if it did, it would have no value. If there is
any value that does have value, it must lie outside the whole sphere
of what happens and is the case. For all that happens and is the case
is accidental... .
6.42 And so it is impossible for there to be propositions of
ethics. .. .6.421 Itis clear that ethics cannot be put into words... .
Ethics and aesthetics are one and the same.
6.422 When an ethical law of the form, ‘Thou shalt...’ is laid
down, one’s first thought is, ‘And what if I do not do it?’ It is clear,
however, that ethics has nothing to do with punishment or reward
in the usual sense of the terms. So the question about the consequen-
ces of our action must be unimportant. At least those consequences
should not be events. . . . There must indeed be some kind of ethical
reward and punishment, but they must reside in the action itself. . . .
34 LUDWIG WITTGENSTEIN

6.4312 Not only is there no guarantee of the temporal immortal-

ity of the human soul, that is to say of its eternal survival after
death, but... is not this eternal life itself as much a riddle as our
present life? ...
6.52 We feel that even when all possible scientific questions have
been answered, the problems of life remain completely untouched.
Of course there are then no questions left, and this itself is the
answer. 6.521 The solution of the problem of life is seen in the
vanishing of the problem.
7. What we cannot speak about we must consign to silence.

A criticism of this passage forms a fitting conclusion for this

section on Wittgenstein because it contains many of the difficul-
ties the positivists must face. There is no need to show that it
conflicts with Christianity, though some items of contrast are
profitable. It is more important to show that if “what we cannot
speak about we must consign to silence,” Wittgenstein should
have spoken less.
In the first place and in the first sentence, it is hard to under-
stand what the term sense means, and how anything can lie
outside the world. One can understand an assertion that no value
exists in the world; but if “the world is all that is the case,” it is the
case that there is no value. On positivistic ground there is no other
world for value to exist in.
Perhaps Wittgenstein was not influenced by Albrecht Ritchl:
he may have independently drawn similar conclusions from his
study of Kant. Kant had indicated a realm of faith beyond the
limits of pure reason—a transcendental world of God, freedom,
immortality, and ethics. Ritch] developed this into a world of
science that had no value, and a world of value that contained no
truth. But at any rate, Ritch] had no qualms about speaking of
values, value-judgments as he called them. Contrariwise,
Wittgenstein said, “It is impossible for there to be propositions of
ethics. ... It is clear that ethics cannot be put into words.”
The difficulty here is not that Wittgenstein has used the word
ethics and implied that it is nonsense. Positivism is justified in
asserting that some terms have no meaning. No one objects to
identifying the words ‘snark’ and ‘boojum’ as nonsense. The diffi-
culty is that Wittgenstein says much more than this: he actually
makes ethical assertions. In 6.422 he gives the ethical proposition,
LUDWIG WITTGENSTEIN ai

“Ethics has nothing to do with reward or punishment in the usual

sense of the terms... [but] there must indeed be some kind of
ethical reward ... in the action itself.” But this Kantian view
cannot serve within logical positivism. For the latter there simply
are no propositions of ethics.
This inconsistency in Wittgenstein can be emphasized by a
contrast with Christianity. The latter threatens eternal punish-
ment for disobedience. Wittgenstein had said, “And what if I do
not” keep the commandment? Christianity replies, you will be
condemned to hell. Now, who speaks the truth, Wittgenstein or
the Bible? Well, if the gentleman wishes to claim truth for his
proposition of ethics, he must renounce positivism. Then, further,
he must defend his proposition that “ethics has nothing to do with
punishment” such as hell. To succeed in his defense, he will have
to retract three fourths of all he had said. Logical positivism must
remain ethically silent. This indeed is what Wittgenstein incon-
sistently does. Denying a future life, and having answered all
possible scientific questions, “the problems of life remain com-
pletely untouched.” Untouched, because “there are no questions
left, and this itself is the answer. The solution of the problem of
life is seen in the vanishing of the problem.” Therefore, murder,
adultery, and theft must be consigned to silence.
 5
Rudolf Carnap

THE motive force behind logical positivism was the twenty-five-

hundred-year failure of philosophy to arrive at any settled con-
clusions. The constant discussions may have varied in some
particulars, but the basic problems were the same and they
remained unsolved. In contrast the positive sciences commanded
respect from the populace and agreement among its practitioners.
The reason for this must be that science had discovered the
proper method and philosophy had not. On this the positivists
were positive.
The differences between the two methods have been stated
with great precision by Rudolf Carnap. There is no doubt that
Rudolf Carnap (1891-1970) was a logical positivist. Surely Russell
was not: he was always dabbling in metaphysical problems; he
even defended some Platonic realities, and wrote serious books on
matters a good positivist would never touch. Wittgenstein was
closer to the strictly orthodox Vienna Circle; but he, too, evinced
interest in nonempirical, ‘nonsayable’ opinions. Carnap, however,
was about as consistent an empiricist as one can hope to find.
Although he began a fruitful literary career in 1928, this section
will summarize only his 1932 article on “The Elimination of
Metaphysics Through Logical Analysis of Language” (in Logical
Positivism, ed. by A. J. Ayer). The reason is, of course, that it
states so clearly the anti-metaphysical arguments.
Through the centuries opponents of metaphysics, he begins,
have charged it with being false, uncertain, or sterile. Modern
logic, Carnap believes, provides a more devastating refutation
because it gives positive results in science and shows negatively
36
RUDOLF CARNAP Bo

that metaphysics is not so much false, uncertain, or sterile, as

meaningless. The rules of language govern the formation of sen-
tences; and therefore pseudo-sentences are such because they
either contain a word that has no meaning or, if the words are all
individually intelligible, they are put together in a counter-
syntactical manner. Both these types of statements occur in
metaphysics.
The reason some words can be meaningless, in spite of the fact
that all words have been introduced into a language for the
purpose of expressing something, is that words change their
meanings in the course of history. In these changes sometimes
words lose their earlier meaning without getting a new one.
Examples would be God, cause, and substance. The meaningful-
ness of a word depends on two factors: (1) its syntax or use must
be fixed by the simplest form in which it can occur, for example, x
is a stone. Then (2) the simplest sentence must satisfy a condition
that can be expressed by four questions: (a) What is this simplest
sentence deducible from, and what statements are deducible
from it? (b) Under what conditions is the sentence true, and
under what, false? (c) How is the sentence verified? (d) What is
the meaning of the sentence?
To illustrate: Arthropodes in a sentence is deducible from ‘x is
an animal, x has asegmented body, and x has jointed legs.’ If any
of these other words—segmented, legs—is to be defined, the regress
stops when we come to an observation or ‘protocol-sentence.’ In
simpler language, the first definition consists in pointing at an
object you can see. Thus the protocol sentence refers to the ‘given’
(das Gegebenes), but, admits Carnap, there is no unanimity on
what is given. Some say ‘sense qualities,’ others ‘total experiences,’
and others say ‘things.’ But we may still claim that a word is
significant only if the sentence is or may be reduced to a protocol
sentence.
Suppose now that some primitive man pointed at a quadrangu-
lar shape, and invented a new word to denote this shape. Then, if a
later philosopher should argue that v-ness always expresses itself
in quadrangular form, but that x-ness, instead of being precisely
quadrangularity, is a hidden quality or force that produces
quadrangularity, Carnap would object that there is no evidence
38 RUDOLF CARNAP

of such a hidden quality and that the word either means quad-
rangularity or nothing at all.
Basically this isthe argument by which logical positivists show
that metaphysics and theology are meaningless. Take the meta-
physical word principle: that is, not an axiom of geometry or a
general law of physics, but the highest principle of the world,
such as Thales’ water or Pythagoras’ number. Under what condi-
tions, Carnap asks, would it be true to say, ‘xz is the principle of y’?
The metaphysician replies: It is true if y arises out of z, or y exists
in virtue of x.’ Now, sometimes an answer of this form is intelligi-
ble, as in a chemical combination or a botanical process. But the
metaphysician does not mean an observable process. When he
says that the world arises out of, or depends on the absolute, he
can give no criterion for testing his assertion, and therefore it is
meaningless. Apxyn indeed had a meaning once upon a time.
Water was at least visible, even if number was not. But though the
word had a meaning long ago, it has now lost it, and though still in
use, is meaningless.

Or take the word God. It has passed through three historical

epochs. In mythology the gods meant visible, physical bodies on
Mt. Olympus. One could go look for them to see whether or not
they were there. In metaphysics God is not the object of any
experience: he could not be pointed at; therefore the word was
meaningless. Theology stands midway between mythology and
metaphysics. Its usage oscillates; and thus either it can be shown
to be false by testing or else it is meaningless. The same analysis
applies also to the absolute, the infinite, being-in-itself, and
similar terms.

In addition to meaningless words, there are also meaningless

sentences. Even when each word in itself has a distinct meaning,
the syntax of a sentence may be impossible; for example, ‘Caesar
is and,’ or, ‘Caesar is a prime number.’ Carnap insists that such
sentences are not false: they are meaningless, for only numbers,
not men, can be either divisible or nondivisible. Many meta-
physical sentences are like this in having intelligible words whose
combination is syntactically impossible. Therefore, grammatical
syntax must be supplemented by a logical syntax.
Carnap then gives a particularly striking example from
RUDOLF CARNAP 39

Heidegger: “What is to be investigated is being only, and nothing

else;. . . nothing, solely being. . . What about this nothing? Does
nothing exist only because the not, i.e. the negation exists?
Or ...does negation exist and the not exist only because the
nothing exists? .. . We assert: the nothing is prior to the not... .”
ete.!
Someone might answer, as Carnap acknowledges, that human
knowledge is limited, and perhaps a higher being knows meta-
physics and can reveal it to us. Carnap’s reply is devastating:
whatever is unintelligible and meaningless cannot become mean-
ingful, however revealed.
Heidegger is but one example, an extreme example; but Hegel
is almost as bad. A great amount of the difficulty, not only in these
two authors, is the transmutation of the copula to be into existence
as a predicate. Kant disposed of this mistake, but people still say
‘I am’ or ‘God is.’
The basic flaw that renders all metaphysics nonsense is the
notion that there is a kind of knowledge inaccessible to empirical
science. There are only three types of meaningful statements, or,
better, only two main types. Tautologies are ipso facto true and
their contradictories ipso facto false; but these ‘analytic’ judg-
ments say nothing about reality. Examples are the propositions
of logic and mathematics. All other meaningful statements are
such because they can be verified by sensory experience. Thus
terms such as vitalism or causality, and value-judgments as in
ethics or aesthetics are all meaningless.
Since on this view only positive science produces truth, the role
of philosophy is that of logical analysis. It eliminates nonsense, as
just explained, and further analyzes and clarifies concepts. The
reason metaphysics has influenced so many people is that it
expresses a person’s general attitudes toward life. But anyone
who designates such a general attitude as a ‘world-and-life view’
“blurs the difference between attitude and theory, a difference
which is of decisive importance for our analysis” (ibid., p. 79). To
express one’s attitude toward life, art is better than metaphysics

“The Elimination of Metaphysics,” in Logical Positivism, ed. by Ayer, p. 69).

40 RUDOLF CARNAP

and more honest: better because art is an adequate expression,

and more honest because metaphysics uses language that gives
the illusion of being theoretical. Metaphysicians are deluded;
poets are not. Of all the arts, music is the best, for it is entirely free
from reference to objects. Metaphysicians are musicians without
musical ability. This condensation of Carnap’s “Elimination of
Metaphysics” is sufficient at this point to characterize the motiva-
tion of logical positivism. For a better understanding of the details
that follow, and especially for a student’s later evaluation of the
movement, two or three theses may be picked out for consideration.
Very basically, logical positivism is an empirical philosophy.
There is no world of Platonic Ideas, noa priori Kantian categories
or innate forms of the mind. So far as mind is concerned, the
tendency is toward behaviorism. But at any rate, truth can be
obtained only through sensory verification. The exact form of the
verification principle varies somewhat from writer to writer; but
all alike depend utterly on sensation.
Worthy of note also is the fact that Carnap’s distinction between
attitude and theory, a distinction he acknowledges to be of decisive
importance for his analysis, has been adopted by several college
textbooks on logic. One of them gives an excellent illustration. A
banker, a lawyer, and a newspaper reporter were discussing
Senator LaFollette. All three agreed that the senator was a
socialist. This is the theoretical truth. The banker and the lawyer
believed that socialism is bad; the reporter thought it good. The
reporter and the lawyer believed that LaFollette was honest and
sincere; the banker thought he was a crook.
This is an interesting illustration, but it is far from proving
that there is any difference at all between attitude and theory.
The distinction in the illustration is that between the historical
statement that LaFollette was a socialist and the evaluative or
ethical statement that he was honest, or dishonest. One is histori-
cal theory, the other is ethical theory.
This criticism, of course, would not impress Carnap. Speaking
as a logical positivist, Carnap is consistent, more or less, in con-
sidering as a theoretical truth the classification of the senator as a
socialist. But the classifications, good, bad, honest, dishonest, are
personal attitudes. Such is logical positivism. Those, however,
RUDOLF CARNAP 41

who reject empiricism, with its observational criterion, reject

also the distinction between truth and attitude, because what the
positivists take for subjective attitude, the others take as true
theoretical propositions. If they are not true theoretical proposi-
tions, then it is nonsense to say that logical positivism is good.
Now, finally, a note on protocol sentences and their ostensive
sources. The beginning of meaning lies in pointing with the
finger at a visible object. Only after pointing can language come
into play. But if we can point at a cat or dog or tree, how can the
meaning of in, for, quickly, greater be seen with the eyes? Not to
mention the square root of minus one?
Such grammatical and mathematical difficulties, plus the
suspicious subjectivism or ethical norms, just might lead to the
demise of logical positivism.
This is a good place to note another factor that could lead to the
demise of logical positivism. In 1932-33 Otto Neurath published
an article in Erkenntniss, the English title of which is Protocol
Sentences.2 This did not seem so dangerous to Neurath when he
wrote it: in fact he thought he was advancing the positivist system.
Nonetheless, he knew he was criticizing Carnap. A few points
from the article are selected here.
Neurath begins by acknowledging that even scientific terms
are not wholly precise, for they are based on protocol sentences,
and “these terms must be vague.” The name ‘Otto,’ in ‘Otto is
observing a thermometer,’ is vague and must eventually be
replaced by mathematical formulas. A scientist, like everyone
else, must begin with ordinary language and purify it by advanc-
ing to physicalistic ordinary language, in which process he will
exclude many proscribed words. Then comes the physicalistic
language of advanced science, free from metaphysical elements.
But the language can be used only for parts of the special sciences.
Other parts of any science book must use some of the lower
language as well. The language of advanced science and ordinary
language coincide primarily in arithmetic. “There is no way of
taking conclusively established protocol sentences as the starting

2In Ayer, ed., Logical Positivism, pp. 199ff.

42 RUDOLF CARNAP

point of the sciences” (ibid.). We can eliminate metaphysics, but

vague linguistic conglomerations always remain.
It is clear that Neurath wishes to eliminate metaphysics as
much as Carnap does. The question is, Can logical positivism do
so? If corrections of Carnap’s view are needed, may not corrections
of Neurath’s view be needed, and so on until the corrected version
is no longer logical positivism?
Neurath has doubts about Carnap’s “primitive protocol lan-
guage,” discussion of which might lead younger men into meta-
physical deviations: “we reject Carnap’s thesis to the effect that
protocol sentences are those ‘which require no verification... .’
No sentence enjoys the noli me tangere which Carnap ordains for
protocol sentences” (ibid., p. 203).
The next subject wi!l be the verification principle, its statement
by A. J. Ayer, and its later modifications.
 6
A. J. Ayer

ALFRED JULES AYER (1910- ) in his Preface to Language,

Truth and Logic (London: Victor Gallanez, 1936) acknowledged
his indebtedness to Russell, Wittgenstein, and Hume. Although
he is closely allied with the logical positivists, and although W. T.
Jones repeatedly classifies him as such, Ayer himself repudiates
the classification and in several places emphasizes or at least
mentions some points of disagreement. That he is in the main
stream of language analysis, however, there can be no doubt. All
propositions are either relations of ideas or matters of facts, as
Hume said; the former are a priori, analytic tautologies, such as
are found in logic and mathematics, and cannot be confuted by
experience—they are simply our “determination to use symbols
in a certain fashion”; whereas the latter are never certain, but
only more or less probable, depending on the degree of verifying
experience. “And in giving an account of the method of validation,
I claim to have explained the nature of truth” (Preface, p. 11).
The Preface itself states Ayer’s basic disagreement with the
logical positivists: “I adopt . . . a modified verification principle.
For I require of an empirical hypothesis, not indeed that it should
be conclusively verifiable, but that some possible sense-
experience should be relevant to the determination of its truth or
falsehood” (p. 11).
At this early point one may reply to an objection that some of his
opponents have made. The strong verification principles, which
these opponents also reject, at least plausibly distinguishes a true
from a false proposition: but this weak verification principle
would allow even a dream to be relevant to the truth of God’s
43
44 A. J. AYER

existence, or other metaphysical propositions equally obnoxious

to Ayer. This objection fails because Ayer insists that there is no
sensory observation that can be relevant to the existence of God
and that theological propositions are devoid of meaning. “If a
putative proposition fails to satisfy this principle [of sensory
verification], and is not a tautology, then I hold that it is meta-
physical . . . neither true nor false butliterally senseless. . . .The
philosopher is not in a position to furnish “speculative truths...
nor yet to pass a priori judgments upon the validity of scien-
tific theories, but . . . his function is to clarify the propositions of
science by exhibiting their logical relationships, and by defining
the symbols which occur in them” (p. 11).
It is to be noted that Ayer’s rejection of metaphysics and
theology depends on a combination of basic empiricism and a
definition of symbols. Concerning a metaphysician Ayer asks,
“Must he not begin, as other men do, with the evidence of his
senses?” (This, of course, is the point at issue, and a rhetorical
question is a petitio.) “What valid process of reasoning can
possibly lead him to the conception of atranscendent reality? ...
No statement which refers to a reality transcending the limits of
all possible sensible experience can possibly have any literal
significance. .. . The fruitlessness of attempting to transcend the
limits of possible sense experience will be deduced, notfrom ...
the actual constitution of the human mind (as Kant did) but from
the rule which determines the literal significance of language”
(p. 19).
If metaphysics is simply bad grammar, one might ask how
Plato, with his great literary skill, came to make metaphysical
assertions. Ayer answers that our language (Greek as well as
English) distinguishes between the thing itself and its various
qualities. Therefore, we say it has these qualities. But the thing is
really the totality of its appearances. Thus the metaphysician is
deceived into thinking that a noun must refer to a thing. The
wrong grammatical assumption is that “to every word or phrase
that can be the grammatical subject of a sentence, there must
somewhere be a real entity corresponding” (p. 35). And with
justifiable sarcasm he refers to Heidegger’s making ‘Nothing’
the name of a reality. Propositions and universals are not real,
and therefore the problems arising from them are fictitious.
A. J. AYER 45

To avoid such blunders, one must realize that philosophy is

chiefly analytic, and never metaphysical. “The philosopher . . . is
not directly concerned with the physical properties of things. He
is concerned only with the way we speak about them.” For
example,

. . amaterial thing cannot be in two places at once. This looks like an

empirical proposition, and is constantly invoked by those who
desire to prove that it is possible for an empirical proposition to be
logically certain. Buta more critical inspection shows that it is not
empirical at all, but linguistic. It simply records the fact that, as
the result of certain verbal conventions, the proposition that two
sense-contents occur in the same visual or tactual sense-field is
incompatible with the proposition that they belong to the same
material thing. .. . There is no logical reason why we should not so
alter our definitions that the sentence ‘A thing cannot be in two
places at once’ comes to express a self-contradiction instead of a
necessary truth (p. 63).

The result is that two marbles cannot be in the same place simply
because we decide to talk that way. There is nothing about
marbles themselves that prevents their interpenetration. Any
time we wish, we can define marbles as bodies that occupy the
same place at the same time.
Ayer is not always so obviously absurd. He makes an interesting
distinction between dictionary definitions and philosophic defi-
nitions. The former are synonyms, by which they assert that one
symbol can be substituted for another. The philosophic definition
of a symbol ‘in use’ occurs “by showing how the statements in
which it significantly occurs can be translated into equivalent
statements which contain neither the definiendum itself nor any
of its synonyms” (p. 68). Here he refers to Russell’s attempt to
explain the meaning of ‘Scott was the author of Waverly.’ The
‘in-use’ definitions bring to light the nonobvious complexity of
ordinary sentences.
A better, a more fundamental application of this type of defini-
tion, aims to show that the is of existence is not the 7s of class
inclusion. He writes: “The ‘is’ which occurs in the sentence, ‘Heis
the author of that book,’” though its form is identical, is not “the
same symbol as the ‘is’ which occurs in the sentence, ‘A cat is a
mammal.’” The reason they are different, says Ayer, is that “the
46 AU AYER

first is equivalent to ‘He and no one else wrote that book,’ and the
second to ‘The class of mammals contains the class of cats” (p. 72).
But the first of these sentences can also be translated, ‘The class
Seott is included in the class author-of-Waverly and the class
author-of-Waverly is included in the class Scott.’ If anyone argues
that Scott and Waverly are not classes, the reply is that a logic
which allows null classes cannot prohibit Sir Walter from being
in aclass by himself. Thus Ayer’s use of a transitive verb, wrote,
instead of the logical copula is certainly not the only analysis
possible, and by many philosophers is considered a worse one.
In chapter 4 Ayer makes it clear that empiricism has no place
for any a priori, and therefore no place for any universal or
necessary proposition, other than tautologies. Arithmetic and
geometry make no assertions concerning the real world or
physical space. Mill had argued that mathematical formulas
were generalizations based on extremely wide inductions, so
wide in fact that there arose the mistaken belief that they were
universally and necessarily true. For Mill they were only highly
probable.
Ayer rejects Mill but not, of course, to accept Kant. There are
indeed a prioris, but nosynthetic a prioris. There are also induc-
tions, but not in Mill’s sense. True, the psychological process of
learning mathematics is inductive, in the sense that the student,
and particularly the originator of new formulae, first makes
many blunders, considers already established truths, and finally
stumbles into something new, correct, and useful. But logically
the formulation does not depend on experience, nor does it give us
any factual information. Mathematics has no empirical verification.
One of Ayer’s examples is a triangle. If we most carefully
measure a visible triangle and find its angles to be a minute
fraction less than 180 degrees, we never count it as a negative
case in induction. “We say that we have measured wrongly, or,
more probably, that the triangle we have been measuring is not
Euclidean” (p. 97). In the next paragraph he continues, “The
same thing applies to the principles of formal logic.” “In other
words, the truths of logic and mathematics are analytic proposi-
tions or tautologies” (p. 100). And here he continues with an
argument against Kant’s a priori synthetic judgments.
Although these analytic principles have no factual content,
A. J. AYER 47

“They do enlighten us by illustrating the way in which we use

certain symbols... calling attention to the implications of a
certain linguistic usage... indicating the convention which
governs our usage of the words ‘if’ and ‘all’” (pp. 104-105).
This view that logic is a convention leads to the conclusion that
“every logical proposition is valid in its own right. Its validity
does not depend on its being incorporated in a system” (p. 108).
Here a difficulty begins to emerge. If each logical principle is so
independent of any system, would it not be possible to assert, not
only contradiction without subalternation, but also subalternation
without contradiction? If each principle of logic is independent of
every other, could we not have universal affirmatives without
contraposition, or vice versa? # might be contradictory of J
without being the contrary of A. And, further, if all these princi-
ples are merely tautologies and conventions, cannot we discard
them all and replace them by other distinctly different ones?
What Ayer says of geometrics surely applies to his view of logics:
“Insofar as they are all free from contradiction, they are all true”
(p. 111). But why must they be free from contradiction? The law
of contradiction is just a convention, and are we not free to
abandon it? If “they simply record our determination to use
words in a certain fashion” (p. 114), why cannot we determine
otherwise? Ayer tries to answer this in his next sentence: “We
cannot deny them without infringing the conventions which are
presupposed by our very denial, and so falling into self-
contradiction.” But does this not make the law of contradiction
something more necessary than a convention? Does not the very
possibility of tautology depend on the law of contradiction? Deny
contradiction—and its obverse, the law of identity—and tautolo-
gies cannot occur. But if this be so, logical principles cannot be
independent of each other: th2y must form a system of obversion,
contraposition, transitivity, and so on.
Ayer often does not treat logic as a mere convention. In his
discussion of the confirmation of a scientific law, he notes that a
negative experiment can be taken to falsify the law, or the law
can be maintained by supposing that the experiment failed to
meet certain requisite conditions, or “that our negative observa-
tion was hallucinatory. And in that case we must [do so and so].
Otherwise we shall be maintaining incompatible hypotheses. And
48 A. J. AYER

this is the one thing we may not do” (pp. 133-134). Yet this is
precisely what we may do, if logic is a mere convention.
Once again, in discussing the alterations of the laws of physics,
he says: “Although we acknowledge that certain standards of
evidence ought always to be observed in the formation of our
beliefs, we do not always observe them. In other words, we are not
always rational. For to be rational is simply to employ a self-
consistent accredited procedure in the formation of all one’s
beliefs” (p. 144).
Is this simply a convention as to how we shall use the word
rational in our conversation? And what about his phrase “ought
always”? How can Ayer’s conventionalism provide a basis for a
universally (at all times and for all persons) obligatory moral
duty? Well, it cannot; for on the next page he adds, “we define a
rational belief as one which is arrived at by the methods which we
now consider reliable. There is no absolute standard of rational-
ity... .Ifin the future we were to adopt different methods, then
beliefs which are now rational might become irrational from the
standpoint of these new methods. But the fact that this is possible
has no bearing on the fact that these beliefs are rational now”
(Dato).
On the contrary, it has a great bearing on the present. No doubt
it is true that alterations in the positive operational laws of physics
would have no bearing on the present standards of rationality. If
a certain equation is the best we can do now, there is no necessary
violation of rationality in later accepting an equation now judged
to be extremely queer. But when Ayer says that there is no
absolute standard of rationality, his words go beyond the present
positive laws of science: they include the law of contradiction, for
this law has traditionally been considered to be the basic test of
rationality. Ayer, of course, insists that present logic is a conven-
tion that may well be replaced in the future. But if I believed now
that next year, or next century, inconsistency and self-contradiction
would be rational, it would have a tremendously destructive
bearing on my belief that these laws are rational now.
A few lines back Ayer imposed on us an “ought always.” This is
a normative demand. To be sure, most people would not regard it
as a moral precept, for present society seems to have restricted
A. J. AYER 49

immorality to murder and thievery. Nevertheless, Ayer has

imposed on us a universal value judgment which, since it is
intended to govern our conduct, can only be classed with morality,
aesthetics, and axiology in general. The question now is whether
or not this theory of ethics allows him to do so.
In chapter 6 Ayer explains his “radical empiricist thesis” as it
bears on ethics, aesthetics, and theology. Many people hold ethical
norms to be genuine synthetic propositions that do not predict the
future course of our sensations, and that therefore ethics cannot
be empirical. To this Ayer replies that insofar as value judgments
are significant, they are ordinary scientific statements of sociology
or psychology; and insofar as they are not scientific, they are not
significant but are only expressions of emotion.
Previous theories of ethics have greatly confused four separate
problems. One problem is the definition of ethical terms, or at
least the possibility of such definitions. For Ayer this and this
alone can constitute a philosophy of ethics. But besides definitions,
writers on ethics have often described moral phenomena. This,
however, is sociology and not ethics. Third, many writers have
included exhortations. But these are not propositions, and cannot
therefore be included in ethics. There is a fourth type of sentences,
namely, “actual ethical judgments.” This fourth division Ayer
hardly explains at all. Without having “yet determined how they
should be classified,” he insists that “they are certainly neither
definitions nor comments upon definitions, nor quotations”; for
which reason “we may say decisively that they do not belong to
ethical philosophy” (p. 151).
Ayer rejects both the subjectivist and the utilitarian reduction
of ethical to nonethical terms. The basis for this rejection is
ordinary English usage. These two ethical theories are both
incorrect analyses of our existing ethical notions. “In our language
statements which contain ethical symbols are not equivalent to
statements which express psychological propositions, or indeed
empirical propositions of any kind” (p. 155).
Here, as frequently, a reader is puzzled by Ayer’s use of
common English to test a theory, when he so strenuously objects
to common English usage in theology.
The absolutists are right in opposing the reduction of ethical to
50 A. J. AYER

empirical concepts. They are right when they say that ethical
concepts are unanalyzable. But their reasons, motives, and
implications are bad. Ethical concepts are unanalyzable because
they are pseudo-concepts. To say ‘you acted wrongly in stealing
the money,’ merely means ‘you stole the money.’ “In adding that
this action is wrong I am not making any further statement about
it. lam simply evincing my moral disapproval of it... . ‘Stealing
money is wrong’... has no factual meaning... expresses no
proposition” (p. 159). Others may have the same or different
feelings about theft: the words are “purely emotive . . . they have
no objective validity whatever.”
In fact, according to Ayer, no one ever disputes about moral
values. What occurs in ethical discussions are disputes about
empirical facts.

When someone disagrees with us .. . we donot attempt to show by

our argument that he has the ‘wrong’ ethical feeling. ... What we
attempt to show is that he is mistaken about the facts in the
case...the agent’s motives... the effects... special circum-
stances. ... As the people with whom we argue have generally
received the same moral education as ourselves . . . our expectation
[that we can convince him by the facts] is usually justified. But if
our opponent happens to have undergone a different process of
moral ‘conditioning’ from ourselves, so that, even when he
acknowledges all the facts, he still disagrees . . . then we abandon
the attempt to win him by argument . . . we finally resort to mere
abuse (pp. 159ff.).

Of course, Ayer means that metaphysicians and theologians

finally resort to abuse. Naturally, nonreligious authors never
resort to abuse.
Well, we gladly acknowledge that Ayer does not resort to abuse;
and we also follow his argument to a point beyond that at which
he stops. For if the condemnation of theft and murder cannot be
rationally justified, Ayer’s objection to them is mere personal,
subjective emotion. Ayer, or at least men older than he, in
England and the United States, were indeed socially conditioned
to have feelings of distaste for murder. But many younger men
and women have not had this moral education and hence have no
feelings against the murder of innocent unborn human beings.
The fairly widespread moral uniformity of the last century or so
A. J. AYER inal

has disappeared. Public education and governmental action have

tended to eradicate Christian morality. But against those who
still oppose murder and advocate capital punishment, Ayer has
no rational argument. He has simply a humanistic emotion.
However, there is something still deeper. Ayer has a feeling
that one “ought always” to avoid self-contradiction. But his emo-
tional view of normative judgments and his logical, tautological,
conventionalism robs his feelings of all authority for anyone else.
The present writer acknowledges that Ayer, quoted in the con-
cluding footnote, does not want logic to be “entirely arbitrary.”
However, even this pious wish would leave the law of contradic-
tion fifty percent arbitrary.
The next step is to show how Ayer, by language analysis, solves
some particular metaphysical problems. Descartes and others
had been troubled concerning their own personal existence, and
more so by the existence of others. A similar, indeed the identical
problem, concerns the existence of a common world. On his phe-
nomenalistic position Ayer agrees that there are no objects whose
existence is indubitable. Even propositions describing the contents
of our sensations are only probably hypotheses. There is indeed a
given, adatum, ein gegebenes, namely, the content of sense expe-
rience; but the description of this content is only probable.
Any question as to whether sense contents are mental or
physical is not an empirical question, but a priori. The realistic
theory with its subject-act-object is not verifiable. A given sense
content is experienced by a particular subject only in the sense
that some sense contents are related to others. There is no sub-
stantial ego or object. To say a sense content exists means merely
that it occurs. It were better to use the term occurrence rather
than existence so as to avoid treating sense contents as if they were
material things. Therefore, sense contents are neither mental or
physical. These two terms can apply only to logical constructions
out of sense contents. A construction may be either physical or
mental, but its elements are neither.
At this point it is hard to resist the temptation of noting that
Ayer’s sense content is as mysterious as Spinoza’s substance. The
latter was both mental and physical; Ayer’s is neither. But what
52 AAV

is a sense content, what is an experience, if not mental? And how

can a logical construction be a physical object?

Well, that question is easily answered: we simply use that kind

of language: “When we refer to an object as a logical construction
out of sense contents, we are not saying that it is actually con-
structed out of these sense contents or that the sense contents are
in any way parts of it, but merely expressing, in a convenient if
somewhat misleading fashion, the syntactical fact that all sen-
tences referring to it are translatable into sensations referring to
them. ... It should be clear also that there is no philosophical
problem concerning the relationship of mind and matter, other
than the linguistic problems of defining certain symbols which
denote logical constructions in terms of symbols which denote
sense-contents” (pp. 190-192).

Bridging the gap between mind and matter is a fictitious

problem, arising out of the senseless metaphysical concept of
substance. “For, roughly speaking, all that we are saying when
we say that the mental state of a person A ata time t isa state of
awareness of a material thing X, is that the sense-experience
which is the element of A occurring at time t contains a sense-
content which is an element of X, and also certain images which
define A’s expectation of the occurrence in suitable circumstances
of certain further elements of X, and that this expectation is
correct: and what we are saying when we assert that a mental
object M and a physical object X are causally connected is that, in
certain conditions, the occurrence of a certain sort of sense-
content, which is an element of M, is a reliable sign of the occur-
rence of acertain sort of sense-content, which is the element of X,
or vice versa” (pp. 192-198).

Now to make some comments, the subjectivism of Berkeley and

the absolute idealism of Hegel, in both of which the fundamental
reality is mind or spirit, have a degree of plausibility. Neither of
these philosophers based his view on mere words or linguistic
usage. Berkeley very pointedly invited us to pull aside the curtain
of words to behold the fair world of ideas. But what can we see in
the terms Ayer has just used: awareness, sense-content, images,
Ae AY HR ae

expectation, and especially, experience? What can we see but

words, words, words?!

‘In 1946 Ayer republished his work with a long Introduction in which he
detailed several changes in his thought. Some of these are trivial replies to still
more trivial objections. Of slightly more importance are a few paragraphs on his
verification principle. By noting that the sense-content necessary to verify an
empirical statement may be any one of a great number of types, and never
precisely one occurrence or one type, he avoids the charge that his principle is too
restrictive. On the other hand, he would not have his principle so little restrictive
that any indicative statement whatever could be taken as meaningful. How he
balances these, it is not necessary to explain here. He also points out that the
verifying statement need not be a part of the meaning of the statement verified.
His example is that blood on my coat can be evidence that I committed a murder,
but blood on my coat is not a part of the definition of murder. None of this greatly
alters his original position, and he continues immediately to repeat that the
existence ofa deity can in no way be verified (p. 15), and is therefore meaningless.
What is of more importance is his acknowledgment that “the word ‘meaning’ is
commonly used in a variety of senses, and I do not wish to deny that in some of
these senses a statement may properly be said to be meaningful even though it is
neither analytic nor empirically verifiable” (p. 15). But this meaning, he insists,
could not be ‘literal’or ‘factual’ meaning. Not only so, but “Furthermore, I suggest
that it is only if it is literally meaningful, in this sense, that a statement can
properly be said to be either true or false” (pp. 15, 16). Perhaps an example might
be that ‘God exists’ means the speaker is stupid.
Of fundamental importance, however, are the next four sentences. After having
insisted that only literal meaning can be true or false, Ayer continues, “Thus,
while I wish the principle of verification itselftobe regarded, not as an empirical
hypothesis [as both Erving and Stace take it to be], but as a definition, it is not
supposed to be entirely arbitrary. It is open to anyone to adopt a different criterion
of meaning... . Nevertheless, I think that, unless it satisfied the principle of
verification, it would not be capable of being understood in the sense in which
either scientific hypotheses or common-sense statements are habitually under-
stood” (p. 16).
The second half of this disjunction does not seem to be the case, and the first half
means merely that a theological doctrine is not a differential equation.
In all, a development of Ayer’s thought is not very evident in this Introduction to
his second edition. But what was not so clear in 1946 became more clear in the
Logical Positivism of 1959. To quote:
“Why should this [verification principle] be accepted [as a convention]? The
most that has been said is that metaphysical statements do not fall into the same
category as the laws oflogic,or as scientific hypotheses, . . . or any other common
sense descriptions of the ‘natural’ world. Surely it does not follow that they [i.e.,
metaphysical statements] are neither true nor false, still less that they are
nonsensical.
“No, it does not follow. Or, rather, it does not follow unless one makes it follow.
The question is whether one thinks the difference between metaphysics and
common sense or scientific statements to be sufficiently sharp for it to be useful to
underline it in this way” (pp. 15-16).

Is not this a total surrender of his original position? Something more along this
line comes in the following section on Feigl.
 Z
Herbert Feigl

HERBERT FEIGL (1902-1976) was a thorough-going logical posi-

tivist who recognized the ineradicable cleavage between his own
view and that which accommodates ethics, religion, and a life
after death. Whether his language is more sympathetic than his
‘attitude,’ it is at least as clear as anyone could wish.
In his Logical Empiricism (Living Schools of Philosophy, ed.
Dagobert Runes[Ames, Iowa: Littlefield, Adams, and Co., 1956],
pp. 325-367) Herbert Feigl wrote:
Probably the most decisive division among philosophical attitudes
is the one between the worldly and the other-worldly types of
thought. Profound differences in personality and temperament
express themselves in the ever changing forms these two kinds of
outlook assume. Very likely there is here an irreconcilable diver-
gence. It goes deeper than disagreement in doctrine; at bottom it is
a difference in basic aim and interest. Countless frustrated discus-
sions and controversies since antiquity testify that logical argument
and empirical evidence are unable to resolve the conflict. In the
last analysis this is so because the very issue of the jurisdictive
power of the appeal to logic and experience (and with it the question
of just what empirical evidence can establish) is at stake.
This is about as clear-sighted and as unbiased as a logical
positivist can be. Perhaps his phrase “The appeal to logic and
experience” betrays a suggestion that whereas logical positivists
use both, theologians use neither. Of course, theologians use logic,
even if they reject experience. The words ‘personality and
temperament,’ however, are unbiased, for they apply equally to
both parties. The term ‘attitude’ may beg the question, for to
modern logicians it often means a nonintellectual, perhaps

54
HERBERT FEIGL 55

emotional, and at any rate noncognitive state of mind, rather

than an intellectual adherence to certain normative propositions.
Feigl’s opponents will repudiate the assumption that there are
differences “deeper than... doctrine,” for “aim and interest”
are doctrine, too. They are not to be sought outside the universe of
truth and falsehood. Nevertheless, no one could be more unbiased
and perceptive than Feig] when he says, “the question of just
what empirical evidence can establish is at stake.”
Feig! is indeed very honest. In the following paragraph he says,
“There will always be those who find this world of ours, as cruel
and deplorable as it may be in some respects [and remember
Feig] had a slight brush with Hitler], an exciting, fascinating
place to livein. .. . And there will always be those who look upon
the universe of experience and nature as an unimportant or
secondary thing in comparison with something more fundamen-
tal and more significant.” Here Feig] does not even exaggerate,
as A. J. Carlson was wont to do in his attacks on revelational
theology. If Feig] made a false step here, it was in suspecting that
the world in some respects is cruel and deplorable. These are
words that have no meaning for the positivists. Ethical terms
have no empirical basis and are therefore nonsense.
Unfortunately, because inconsistently, Feig] further contrasts
men and movements that have “respect for the facts of experience,
and openmindedness” with others who are “more impatient...
and tender-minded” (p. 326). The sentiment is polite and gentle
enough; but his theory makes such evaluations impossible.
However, he must be allowed to get on with his argument.
In fact, he is harder on those who more or less agree with him
than he ison metaphysicians and theologians. Some on his side of
the great divide are radical reductionists: Organisms are nothing
but machines; mind is nothing but matter; matter is nothing but a
cluster of sensations; good and evil are nothing but projections of
our likes and dislikes. Feigl’s logical empiricism considers all
these to be “reductive fallacies.” He will reject ‘nothing but,’ as
well as ‘something more,’ and accept ‘what is what.’ Artificial
reductions and wishful thinking must both go (pp. 326-327).

On this principle philosophy asks two chief questions: What do

you mean? and How do you know? Answering these questions will
56 HERBERT FEIGL

indeed aid scientifie sociology and politics, but the primary aim
of philosophy is not the construction of a world-view nor a vision
of a way of life. Philosophy concerns itself with consistency,
testability, adequacy, precision, and objectivity. This concern has
already rid us of magical, animistic, and mythological explana-
tions; the remains of metaphysics and theology must now be
swept away.

These unsuccessful methods thrive on verbal confusion. There-

fore, phraseology about ‘absolute space’ and ‘absolute time’ must
be shown to be devoid of factual meaning. The behaviorists have
already alerted us that the scientific content of psychology can be
formulated in physical language so that mentalistic terminology
is an illusion. Further progress in every science will similarly
depend on distinguishing the functions of language.
Language has six types of meaning, divided in two groups of
three each: Cognitive meanings are purely formal, logico-
arithmetical, or factual; and noncognitive meanings are pictorial,
emotional, or volitional. Many metaphysical theories result from
the erroneous presumption that factual meanings are present in
purely emotive appeals or in formally correct grammatical
structure. Logical empiricism is primarily concerned with cogni-
tive meaning. The meaning of words consists in the way they are
used. The definition of a word amounts to a statement of the rule
according to which we decide to employ the term. Finally all
terms must be reduced to a small number of basic terms, and
these must be immediately connected with experience. The
terminal step, therefore, is ostensive definition.
The purely formal terms of logic and mathematics, of course,
have no ostensive definition. They are defined only by relating
them to each other. Therefore, logic and mathematics have no
factual content. They are pure symbolism.

The important thing is to understand the criterion of factual

meaningfulness. Reference to experience, the ostensive element,
is the key. “A sentence is factually meaningful only if we are in
principle capable of recognizing such states of affairs as would
either validate or invalidate the sentence. If we cannot possibly
conceive of what would have to be the case in order to confirm or
disconfirm an assertion we would not be able to distinguish
HERBERT FEIGL 57

between its truth and its falsity” (p. 334). There must bea differ-
ence capable of observational testing.
The term metaphysics is elastic, and all sorts of problems have
been discussed under its name. They are not all meaningless.
Speculative cosmology, derived by extrapolation from observa-
tional evidence, can be factually meaningful. Examples are the
heat-death of the universe, the origin of life, and the future path
of evolution. But though not necessarily meaningless, these
matters are so vague and uncertain that the guesses are apt to
remain barren and can therefore be considered as the risky,
disreputable extreme of science.
Deductive, rationalistic metaphysics, however, is completely
devoid of factual meaning. Intuitive metaphysics confuses having
an experience with knowing something about it. Transcendental
metaphysics, an attempt to uncover the basic categories of reality,
or speculations about the ‘absolute’ generally contain an ample
measure of ‘absolutely’ untestable pseudo-propositions. Real
science has progressed beyond the ideas of absolute time and
space, substance, numbers as real entities, causes and effects,
vital forces and entelechies. All these are meaningless.
After the question, What do we mean? comes the question, How
do we know? How can a truth-claim be justified on the basis of
observation? This is not a question of psychology. Psychology
itself is a science, a system of truth-claims, and therefore stands
in need ofjustification. Epistemology, on the other hand, has todo
not with origin and temporal development, but with logical
structure and empirical validation.
It is possible to oversimplify the observational criterion of
truth. Very few statements can be validated by direct observation.
Most of knowledge is very indirect. This constitutes the problem
of justifying the principle of induction and probability. It is a
difficult problem. “All attempts to ‘justify’ inductive inference on
rational, empirical, intuitive, or probablistic grounds have turned
out to be utter failures” (p. 341). Induction and probability must
assume that the observations give fair samples of the phenomenon
under study. Yet this assumption has no empirical basis.
Feigl offers a startling solution to this problem. “Logical
Empiricism cuts the Gordian Knot by bluntly asking the question,
58 HERBERT FEIGL

‘What can justification possibly mean here?’” There are only two
conceivable answers: deductive proof or inductive evidence. “The
‘yreat problem of induction’ therefore, consisted in the impossible
demand to justify the very principles of all justification” (p. 341).
The principle of induction, to be sure, is nota piece of knowledge,
but a rule of procedure. It is therefore a tautology. Thus we are
able to escape Hume’s skepticism and all animal faith. “The
procedure of induction therefore, far from being irrational,
defines the very essence of rationality” (p. 342).
At this point a professor of philosophy cannot resist the tempta-
tion to change from exposition to criticism. One criticism is: if
evidence is identical to justification, all generalizations are ipso
facto justified. A person who had studied American history only
so far as 1959 could, in 1976, conclude that no American president
has ever been a Roman Catholic. This is, of course, just another
example of a white crow. No doubt it is true to say that no
president before 1960 was a Catholic; but this is a ‘complete’
induction, that is, a deduction, such as never occurs in science.
Science always operates by means of incomplete induction; and
clearly these do not justify universal propositions.
Then, too, if induction is a tautology and a rule of procedure,
someone other than Feigl has an equal right to choose another
tautology and a different rule of procedure. There is no observa-
tional, ostensive, or factual compulsion to adopt one rule rather
than another. They are all purely formal and stand as equals
before the judge. If, indeed, Feigl’s criterion produces certain
results, another person who has no interest in those results can
reject results and criterion together. What is more, the results
may be accepted without accepting the premises, for the same
conclusion may be deduced from differing sets of premises. In
Yosemite the ranger will show the tourist a series of clay models
exhibiting the stages in the formation of Half Dome and the
Merced Valley. Now, if conditions were such as those depicted in
the first model, and if the processes were such as those depicted in
the succeeding models, the result might indeed be Half Dome.
But Half Dome does not necessitate either the processes or the
original topography. Thus Feigl’s choice of induction remains
Feigl’s choice; and for the previous reason it is a bad choice.
Logical empiricism aims to rid us of metaphysical absurdities
HERBERT FEIGL 59

and meaningless terms. That there are such and that we should
rid ourselves of them is beyond question. But whether the partic-
ular language analysis of logical empiricism succeeds is a differ-
ent question. Feig] next mentions the traditional issues of ‘the
reality of the external world’ and ‘the existence of other minds.’
When phenomenalists and subjective idealists appeal to experience,
Feig] stresses how vague the word experience is. One would think
that logical empiricism would have an exceptionally clear concept
of experience. Feig] then dismisses subjective idealism with the
rhetorical question, somewhat reminiscent of Dr. Johnson, “Any
need to emphasize how absurd that is?” “Metaphysical realism”
fares no better. But if so, how can “Empirical Realism” avoid
some impossibilities? Was not modern metaphysical realism
empirical? Feig! replies, “The term ‘real’ is employed in a clear
sense and usually with good reason in daily life and science to
designate that which is located in space-time and is a link in the
chains of causal relations. .. . The reality, in this sense, of rocks
and trees, of stars and atoms, of radiations and forces, of human
minds and social groups, of historical events and economic pro-
cesses, is capable of empirical test” (p. 342).
The paragraph adorned with this fine literary sentence merits
scrutiny. First, it is doubtful that the term real is clearly con-
ceived in daily life, or even in science. Rather it seems like a word
which language analysis might exclude as meaningless. As for
daily life, a billion people, Christians, Moslems, Jews, and some
others, regularly say that God is real. Feig] can maintain that this
is meaningless; but he cannot deny that it is common usage.
Therefore, daily life does not restrict the term real to space-time
objects of experience. Furthermore, Feig! should not attribute to
it any position as a link in the chain of causal relations; for only a
few pages back he had classified cause-effect relations with
several other “obscure faculties and mythical powers [which]
have gradually disappeared from respectable science” (p. 338).
Then, too, a good section of ‘daily life’ is not apt to assert that the
human mind is extended in space, for most people are not behav-
iorists. But perhaps the most bothersome point in the paragraph
is his acceptance of the thesis, “The only meaningful way to speak
about things is in terms of what they are knowableas.” The thesis
itself may be true: it is inherent not only in Berkeleyan idealism,
60 HERBERT FEIGL

but also in Cartesianism (which is not “replete with pictorial and

emotional appeals”). But though the thesis be true, it hardly
accords with the rest of what Feigl says. On an empirical basis
the thesis results in idealism; on a Platonic basis it results in the
rational realism of suprasensible ideas. These two, however, are
precisely what Feig] rejects.
From among the several other matters that Feig] discusses the
present study will conclude with two: logic and ethics—in reverse
order. The previous section on Wittgenstein exposed some of his
ethical difficulties. Let us now see what Feigl can do.
The term ought is irreducible and is the irreducible directive
component of moral value judgments. Moses would hardly have
agreed with this; but we may accept it as the position of logical
positivism. Feig] explicates: “An ethical imperative like the
Golden Rule simply means: ‘Would that everybody behaved’”
thus and thus. “This sentence, having its accent in the emotive
appeal, could not possibly be deduced from a knowledge of facts
only; itis neither true nor false. ...[Any other view] dogmatically
proclaimed or merely abstractly assumed... involves confusions”
(pp. 354-355).
Of course, there can be a factual, empirical, decidable debate
as to whether certain means will produce certain ends. But as
Dewey said, there is no final end.! Like Dewey, Feigl confines
ethics to the use of “leading standards, thoroughly empirical, to
be sure, in the light of which we evaluate the mutual adjustment
and harmonization of ends and means.” But the choice of an end,
it is clear, remains a meaningless emotive device for the direction
of attitudes. “The ever present possibility of asking the question,
‘But is this really good?’ shows that no descriptively delimited
locus of valuableness forces its acceptance upon us as an ultimate
criterion” (p. 356). Any other opinion “manifests a not fully liber-
ated, pre-scientific type of mind.”
But suppose I choose as my end the cultivation of a not fully
liberated pre-scientific type of mind? Feigl wants “no other
standards than those prescribed by human nature and by our

‘Cf. Gordon H. Clark, Dewey (Phillipsburg, N.J.: Presbyterian and Reformed,

1960, 1978).
HERBERT FEIGL 61

own insights in the possibilities of improving human nature”

(p. 356). But my human nature and insight shows that the pres-
ervation of snail-darters is better than any TVA power station on
the Tennessee River. That is why the savage, pre-scientific men-
tality of the thirteenth century Highlanders in Scotland is so
superior to the degenerate minds of logical positivists.
Feigl can fulminate all he wants: “a truly empirical study of
human nature and social conduct discloses a considerable com-
mon denominator in at least the basic needs of all individuals
living in the context of cooperation and mutual dependence”
(p. 356). Well, hardly. First, there is no empirical test of a “need.”
Second, the ideals and desires of Buddhists, Hindus, Moslems,
Christians, and logical positivists are utterly incompatible. They
have no common ethical principles at all. Did not Feig! himself in
his opening paragraph acknowledge the “irreconcilable diver-
gence” between secularists and religionists? Nonetheless, here,
thirty pages later, he suggests that we ought (?) to develop
“genuinely kind and altruistic attitudes.” But this is only his
emotional, prescientific attitude. And while a proper moral
attitude must be prescientific, not every prescientific attitude is
moral—altruistic attitudes, for instance.
Feig! himself half acknowledges this, for instead of ascertain-
ing inductively the majority desire of all people, he requires a
logical positivistic elite to impose its views of humanity on the
backward prescientifics. Thus, the methods of torture used by
the Communists “may nevertheless be justified on the basis of the
expected results of the new measures for the totality of mankind”
(p. 357). Heaven preserve us!
The final point of criticism is Feigl’s view of logic. If there is
one thesis that permeates logical positivism it is the purely formal,
analytical, tautological character of logical and mathematical
propositions. Otherwise we would have to soar into the Platonic
empyrean, like a Kantian dove beyond the resistance of air where
progress is illusory. Says Feigl:
The advocates of the unity of science . . . emphasize the distinction
between the formal sciences (logic and mathematics) and the
factual sciences. [In logic] the main progress . . . has depended
on... theelaboration of the symbolic machinery of mathematical
logic and the semantical and syntactical analyses of its meaning
62 HERBERT FEIGL

and structure. ... The rules of deduction belong to the internal

regulative mechanism of a consistent language. . . . Logical rules
thus guide us in the transition from premises to conclusions. . . .
The theorems or laws of logic are analytic sentences. . . . The law of
non-contradiction, for example, is inescapably and infallibly true
as long as we agree to mean by a ‘sentence’ an expression which is
either true or false... . (pp. 345-346).
So far, all is clear. The laws of logic are analytic tautologies.
But immediately, to avoid adifficulty, Feig] makes an astounding
admission that measures nine onthe positivists’ Richter scale:
This view of the nature of logical laws has been criticized as
conventionalistic. It is said to assert that logical laws are a matter
of arbitrary decree concerning the use of symbols. Obviously
enough [and here Feig! admits the criticism] from a purely formal
(syntactical) point of view a system of logic is just one calculus
among an indefinite number of others. And yet [as the first tremor
announces the imminent earthquake] we cannot speak of alternate
logics in exactly the same sense in which we speak of alternate
geometries. The uniqueness of logic seems to depend on its purpose
in the use of language; as long as we wish our language to use
unambiguous and consistent designation rules [i.e., as long as we
wish to speak intelligently and intelligibly] we simply must have
{the law of non-contradiction].
With this must logical positivism collapses into debris, and in the
settling dust we read “Whatever calculus...we may find
adequate for this or that scientific purpose, our determination to
employ symbols with constant meanings necessitates [ital. mine]
the retention of a yes-or-no logic somewhere as, so to speak, the
ultimate court of appeal” (p. 347). Through this dust, this ‘deter-
mination,’ this ‘necessitates’ and this ‘yes-or-no,’ we can see the
image of God in man, the Logos who enlightens every man who
comes into the world. Logical positivism cannot possibly be
consistent.
 8
The Later Wittgenstein
THE Philosophical Investigations (tr. G. E. M. Ansombe [New
York: Macmillan, 1953]), the Blue and Brown Books, as well as
the less technical parts of the Philosophical Grammar, are as
impossible to summarize as the Tractatus. There is also a diffi-
culty in evaluating them, but a difficulty of a different sort. The
Tractatus can, without much controversy, be adjudged profound;
but the Investigations and the Blue and Brown Books often give
the impression of triviality. This harsh judgment may anger
some devotees of Wittgenstein, but it is not beyond possibility
that Wittgenstein himself might have agreed. The reason is that
in these volumes he investigates numerous phenomena and quirks
of language. Nearly all are interesting; but it is hard to see that
all are profound. Even some that are suggestive seem to lead
nowhere. And the author admits as much.
The Preface of the Philosophical Investigations states, “After
several unsuccessful attempts to weld my results together into
such a whole, I realized that I should never succeed. The best that
I could write would never be more than philosophical remarks;
my thoughts were soon crippled if I tried to force them on in any
single direction against their natural inclination” (p. ix).
Evidence of this comes in the first few pages. Wittgenstein
begins with a quotation from Augustine. It would be well to
consider this quotation simply as a theory of language Wittgen-
stein wishes to discuss, rather than as Augustine’s theory.! The

\Wittgenstein does not begin at the beginning of a sentence. He omits “Prensa-

bam memoria.” More seriously, he passed over a previous “sed ego ipse mente

63
64 THE LATER WITTGENSTEIN

theory is a very crude one, which Wittgenstein expresses thus:

“the individual words in language name objects . . . every word
has a meaning. The meaning is correlated with the word. It is the
object for which the word stands.” Embedded in this description
there seems to be the assumption that the “object for which the
word stands” is a physical object, such as a rock or tree. Certainly
this is far from the view of Augustine, who was not “thinking
primarily of nouns like ‘table,’ ‘chair,’ ‘bread,’ and of people’s
names...” (§10). There is no objection, however, to discussing
the theory itself, divorced from any historical connection.
Wittgenstein now imagines a primitive language suitable toa
builder and his assistant. The language contains words such as
stones, beams, slabs, and not much else: “Conceive this as a
complete primitive language,” says Wittgenstein (§2). The builder
yells, ‘Slab!’ and his assistant brings him one. In a moment
Wittgenstein will ask whether ‘Slab!’ is a word or a sentence.
Awaiting his answer, one may surmise that if ‘Slab’is a word, itis
not evena word. A language of three nouns is not a language. The
sounds would have no meaning. At the very least, it would be
necessary to say, ‘That thing there is a slab.’ But if ‘Slab!’ is a
sentence, as it must be, an elliptical sentence, then we could never
“conceive this as a complete primitive language.” Even the most
primitive language would have, not only other imperatives, but
also several indicative verbs. Too often Wittgenstein proceeds on
a premise the reader cannot accept.
A page or so later Wittgenstein expands this language to cover
the needs of a shop-keeper (§8). There will be numerals, the
names of some colors, and, what is fatal to the original complete
language, the words there, this, and 7s.
A reader is most likely to say to himself, these are not two
different languages; in fact, neither one is a language; they are
parts of a more inclusive language. But Wittgenstein answers,
“Do not be troubled by the fact that languages (2) and (8) consist
only of orders. If you want to say that this shows them to be
incomplete, ask yourself whether our language is complete;—

quam dedisti mihi, deus meus, cum . . . edere vellum sensa cordis mei. . . .” Nor
does Wittgenstein refer to the tractate de magistro.
THE LATER WITTGENSTEIN 65

whether it was so before the symbolism of chemistry and the

notation of infinitesimal calculus were incorporated init. . . . Itis
easy to imagine a language consisting only of orders and reports
in battle” (§§18, 19).
Now, first, this does not seem easy at all. Of course, no one
denies that a language can increase its vocabulary, and not only
by calculus and chemistry. Words also drop out of usage. In this
sense no language is ever complete. But can anyone realistically
imagine human beings restricting their conversation to orders
and reports in battle, without words for rainy and sunny weather,
agriculture, music, love, elementary politics and economics? No
human group can exist without these, however primitive other-
wise its form may be. Even the ‘language’ of battle must include
words that apply without reference to battle. It would include not
only this and there, but also chemical, physical, and engineering
terms useful in other universes of discourse. Wittgenstein himself
admits, “To imagine a language means to imagine a form of life.”
But who can imagine a form of life restricted to battle commands
or building directions? On the same page he continues, “Is the call
‘Slab’ in example (2) a sentence or a word?—If a word, surely it
has not the same meaning as the like sounding word of our
ordinary language, for in §2 it is a call. But if a sentence, it is
surely not the elliptical sentence: ‘Slab!’ of our language” (§19).
None of this is convincing. Actually it seems prima facie false.
That the meaning of verbal signs depends on their usage may
indeed be true. But if so, the “call” Slab is surely the elliptical
sentence of our language. Why not? What else could it be? A mere
word, all by itself, without any implied context, has no meaning
and is not a part of ordinary usage. The nearest thing to an
isolated word is some sentence like ‘Slab is a word.’ And now we
are into medieval first and second intensions.
There is another difficulty in proceeding into the Investigations.
Clearly Wittgenstein had altered his views considerably between
1920 and 1950. Therefore, there is a possibility, at least a point to
be examined, of inconsistency in the later publications. Discard-
ing various of his old ideas, he may have retained some that
should also have been deleted. There was indeed change. Whereas
the earlier theory asserted a one to one relation between each
word and its referent, the later Wittgenstein acknowledges that
66 THE LATER WITTGENSTEIN

this is an illusion. Certainly he rejects the substitution of an

artificial language for ordinary language. The point is important
and his argument is devastating to earlier positivism. The /nves-
tigations, §120 says, “When I talk about language (words, senten-
ces, etc.) Imust speak the language of every day. Is this language
somehow too coarse and material for what we want to say? Then
how is another one to be constructed? And how strange that we
should be able to do anything at all with the one we have?” Then
he continues with a hypothetical conversation in reply to an
objector.
Whether therefore, or simply in addition, the later Wittgenstein
rejected the picture theory of language—at least some of his
paragraphs seem to reject it. Anthony Kenny notes three alleged
contrasts between the earlier and the later Wittgenstein.? The
first is the metaphysical atomism of the Tractatus. Kenny agrees
that this was indeed discarded. Second is the change from
emphasis on the formal structures of logic to the study of the
idioms of ordinary language. Kenny asserts that this is partly
accurate and partly misleading. The third, the alleged rejection
of the picture theory of meaning is “almost wholly misleading.”
This point deserves examination.
Certainly in the Tractatus, for example from 3.5 to 4.021,
thoughts and propositions are pictures of facts. One of his well
known examples was the reconstruction of an auto accident in ~
court by placing toy autos and dolls in the same relationships the
cars and persons had on the highways. Various critics (Erik
Stenius, Max Black, Anthony Kenny) have discovered profound
difficulties in this theory. Even on a more superficial level the
picture theory lacks clarity. Musical notation on a sheet of paper
can neither look like a sound nor sound like a sound. The Tractatus
(e.g., 4.014-4.023) does not succeed in showing what isa picture of
what. At the moment, however, the point is not the stupidity of
the picture theory, but the question whether or not Wittgenstein,
in altering it, altered it enough to get rid of it entirely. That he did
not quite succeed is clear from the following paragraphs. “To
have understood the definition means to have in one’s mind an

*Kenny, Wittgenstein (Cambridge, Mass.: Harvard University Press, 1973),

pp. 219-220.
THE LATER WITTGENSTEIN 67

idea of the thing defined, and that is a sample or picture” (§73).

“How am I to imagine this mechanism . . . perhaps a drawing
reduced in scale may serve . . . it simply invites me to apply the
picture I am given. A picture is conjured up which seems to fix
the sense unambiguously.... The picture should be taken
seriously. ... We have a vivid picture . . .” (§§425-427). In some
paragraphs, for example, 376, 382, he seems to be criticizing the
picture theory. No doubt he sees some difficulties in it; yet he does
not totally reject it. Therefore, this remnant from his earlier
position allows some basic inconsistencies to blur the thrust of the
Investigations.
The thrust, or the new analogy to replace the picture theory, is
the notion of language games. Words are like chess men, playing
cards, tennis and baseballs, or darts: they mean what they are
used for. A word functions like a pawn or rook in chess. It is not
necessary that the chess men be carved in the Staunton design.
Given that the rook does not look like a knight, the pieces can be
any shape no matter what. They are what they are because of the
rules governing their moves.
The idea that languages are games is totally destructive of
logical positivism. Hordern will later insist that the religious
language game can be played as well as the scientific language
game. There is no compulsion to play tennis rather than soccer;
and the positivist who chooses science has no basis to deny a
religionist his right to play a different game. This argument may
not be very complimentary to religion, nor to science either; but it
is devastating to positivism. The more fundamental question,
however, remains. Is language just a game?
The Tractatus makes no mention of language games. The later
writings are full of it. As early as 1930 Wittgenstein compared
language with chess. Now, as the physical shape of the pawn is of
no.importance, so also rules and usage determine the meaning of
the symbols in mathematics, and the variables in logic. The
difference between syntax and chess lies solely in their applica-
tion. Die Philosophische Grammatik, begun in 1932 or therea-
bouts, constantly revised, and only published posthumously in
1969, devotes a full chapter to the analogies between chess and
arithmetic. These anticipations are worked out much more fully
later on.
68 THE LATER WITTGENSTEIN

The working-out, however, is beset with puzzling remarks. In

the Investigations, §27, he says, “In languages (2) and (8) there
was no such thing as asking something’s name. This, with its
correlate ostensive definition, is, we might say, a language game
on its own.”
Parenthetically, we might also say not. “... there is also a
language game of inventing a name for something, and hence of
saying, ‘This is...’ and then using the new name. (Thus, for
example, children give names to their dolls and then talk about
them and to them. Think in this connection how singular is the
use of a person’s name to call him!)”
What is so singular? If we yell, ‘Hey, come here,’ and no one
other than Jones is in the vicinity, he understands that we are
calling him. But if there are a dozen workmen on the job, no one
would know whom the boss is calling. He would have to yell, ‘Hey,
Jones, come here.’ Is this so singular, perplexing, or profound?
Similarly, in many places Wittgenstein’s readers can only wonder,
what is the point?
But to return to the main notion of language games. Wittgenstein
examines the term games. “Someone might object against me,
“You take the easy way out. You talk about all sorts of language-
games, but have nowhere said what the essence of a language-
game, and hence of language, is: what is common to all these
activities....’ This is true,” replies Wittgenstein. “Instead of
producing something common to all that we call language, Iam
saying that these phenomena have no one thing incommon...
but that they are related to one another in many different ways.
And it is because of this relationship, or these relationships, that
we call them all ‘language’” (Philosophical Investigations, §65).
To make this quotation applicable, one would have to specify
the relationships, and these might turn out to be the definition of
language. Inasmuch as cat is related to cactus, because both
words begin with C, Wittgenstein’s brief reference to “relation-
ships” is not encouraging. Or if the example of cat and cactus is
deemed too trivial and irresponsible, politics and religion may be
substituted. Politics is one language game; religion is another
language game; politics is often related to religion. But does this
help to define politics or religion? Do we call Christianity or
Islam a religion because it is somehow related to politics? In fact,
THE LATER WITTGENSTEIN 69

if language is difficult to define, still the similarities between

Chinese and German are more easily found than the similarities
among Islam, Hinduism, and Christianity.
If now the relationships among the several languages could be
specified, would not they themselves be the common quality?
Wittgenstein furnishes an explanation: “Consider . . . ‘games.’
I mean board games, card games, ball games, Olympic games,
and soon. What is common to them all? Don’t say, ‘There must be
something common, or they would not be called ‘games’—but look
and see whether there is anything common to all.—For if you look
at them you will not see something that is common to all, but
similarities, relationships, and a whole series of them at that”
(§66).
Wittgenstein has chosen a good example of something difficult
to define. Football and bridge both have opposing teams; these
teams play to win. But in trying to show how difficult it is to
define ‘game,’ Wittgenstein uses the example of achild throwing
a ball against the wall.
Here there are neither rules, competitors, nor winning and
losing. Similarly, for Wittgenstein language cannot be defined.
Buta difficulty emerges. Perhaps we could both admit that the
child was amusing himself and that this amusement was not a
‘game.’ In this case ‘rules’ would become a common quality, a part
of the sought-for definitions. Or the situation could be analyzed in
a different way. For a somewhat older child, throwing a ball
against the wall, playing catch, or putting balls through the
basket could be regarded, not asa game by itself, but as practice
for an organized game. Spring training may not be a baseball
game in itself, but its. skills are used in a game. So, too, a child’s
prattling may be, not so much one language game as distinct
from another, but simply an elementary practicing of the adult
usage. More profoundly the language of botany should not be
regarded as a language of its own—though the botanical terms do
not occur in baseball or in the stockmarket—nor has politics a

30n the definition of religion, see chapter one of the writer’s Religion, Reason,
and Revelation (Nutley, N.J.: Presbyterian and Reformed, 1961).
70 THE LATER WITTGENSTEIN

language of its own. These are two different subjects of conversa-

tion, each with some terms of its own; but the one language, with
the indispensable words is, the, for, because, not, when, all, some
nonsense, ete., is ordinary English. Not restrictedly English
either. German is the same ‘game’; the rules are the same, only
the balls have a different color.
Wittgenstein wants to hold that the term game cannot be
defined, and that therefore language games have nothing in
common. To quote: “If someone wished to say: “There is something
common to all these constructions, namely the disjunction of all
their common properties’—I should reply: Now you are only
playing with words. One might as well say: ‘Something runs
through the whole thread—namely the continuous overlapping of
those fibres” (ibid. §67).
This, however, does not seem so ridiculous as Wittgenstein
intends, unless one of us has lost the thread of the discourse. It is
precisely because of the overlapping of the fibers, in such a way
that they cannot be easily pulled apart, that a thread is distin-
guished from loose fibers that are not thread. But more to the
point, his analogy is basically poor. When he says, “The strength
of the thread does not reside in the fact that some one fibre runs
through its whole length,” the blunt materialism or at least cor-
porealism is evident, even when he adds, “but in the overlapping
of many fibres.” A definition of ‘game,’ or even of one game,
tennis, does not depend on a corporeal continuum. In tennis the
balls may change; in baseball it hardly ever happens that a single
ball survives an inning. Hence, at this point Wittgenstein’s
attempt to prove that language cannot be defined, fails.
As an anticipation of how the present analysis will continue, let
it be said that insofar as Wittgenstein’s later criticisms of meta-
physics depend on the indefinability of games, they have little
force. However, he keeps on in his interesting way and enumer-
ates more peculiarities in language than an ordinary person
could think of in a long time. His aim is to show that neither
games nor languages can be defined. The varieties in both these
activities have no common quality. Yet the multitude of his
observations do not quite guarantee his conclusions.
Consider games again. One may surmise that originally a game
was, not just rest from labor, but yet some form of active recreation.
THE LATER WITTGENSTEIN ral

The ploughman worked hard all day and then in the evening he
amused himself in some artificial problem. These amusements
became more structured and brought in one or more other
players. After civilization had further developed, games became
work and professionals commanded high salaries. Today for the
American spectator the deliberate violence of hockey and football
remains a game, as does also the cruelty of bull fights in Spain
and Latin America. But for the participants they are work. The
players are actors on a stage, and when actors present a play, they
are not playing.
This is why one of Wittgenstein’s unwitting assumptions must
be questioned. “How should we explain to someone what a game
is? I imagine that we should describe games to him” (ibid. §69).
This is probably so, if we are talking to asmall child. But for an
adult, it is wrong. Two related sources of ambiguity require
recognition. First, the historical point at which games become
work escapes notice. Second, this historical point includes a
change in the motivation. It is the motivation of the participant,
not his actions, that makes the motions fun or work. Football or
tennis may be fun for the high schooler. For the professional the
same motions are work: such hard work and such strong competi-
tion that the football idol and the Olympic contestants, and the
horse that wins the Derby, often must use drugs. The amateur or
duffer plays tennis according to the rules the Wimbleton champions
use. The rules are indeed the similarity that makes tennis tennis;
but they do not make tennis a game. Different rules make chess
chess. But it is a game because of purpose and motivation. Sim-
ilarly, different vocabularies make German German, and English
English. Lesser differences in vocabulary make botany one
subject and politics another. But none of these prevents them
from being ordinary language, and just one ‘game,’ defined by
the rules of logic.
‘The notion that we speak several languages, even when it is all
English, as we play several games each with different rules—and
this seems to be the main notion throughout the Philosophical
Investigations—depends on a view of logic which cannot be sus-
tained. In §81 Wittgenstein writes, “F’. P. Ramsey once empha-
sized ... that logic was a ‘normative science.’ I do not know
exactly what he had in mind, but it was doubtless ... that in
T2 THE LATER WITTGENSTEIN

philosophy we often compare the use of words with games and

calculi which have fixed rules, but cannot say that someone who
is using language must be playing such a game . . as if it took
the logician to show people at last what a correct sentence looked
like.”
Then, after describing some boys amusing themselves by
tossing a ball around without any rules, he adds a game in which
the boys make up the rules as they go along, and later alter some.
These remarks, in my opinion, seem to confuse the laws of logic
with the differences between and the changes within English and
German. Perhaps some day French will adopt the indeclinable
masculine form for all predicate adjectives, as German has done;
and the process of replacing strong verbs in English with weak
verbs will doubtless continue. But this is not the case with the
laws of logic. Wittgenstein’s analogy is even weaker than the
grammatical changes in the history of a language.
A concrete application of paragraphs 84 and 85 would be the
observation that the rules of chess do not determine the opening
move, much less the next ten. Similarly, the rules of language and
the normative laws of logic do not determine whether one dis-
cusses botany or politics. But none of this seems to have any point.
Grammar requires a plural subject to have a plural verb; yet
grammar remains grammar, and English, English. Likewise,
the normative laws of logic are normative, even though they do
not limit the subject matter. Regardless of subject matter,
however, violations of grammar can come close to destroying
English. In the history of Greek the purposive conjunction hina
became na, the preposition to used with an infinitive. This caused
little confusion. The invention and introduction of the word its
actually improved English. But if too many grammatical rules
are broken, the result is a different language, as Latin became
French, Spanish, and Italian. If, however, the logical rules are
broken the result is nonsense, no language at all.
Wittgenstein’s remarks here are far from invalidating the
notion that logic is a ‘normative science.’ Further to support his
contention, he spends some interesting pages exhibiting the (non-
Wellhausenian) difficulties of determining who Moses was. But
however difficult it might be to determine the definition of Moses,
no one can be so perverse as to say that Moses was Joshua or
THE LATER WITTGENSTEIN as

Solomon; yet Wittgenstein’s theory of language games seems to

require as much: identifying Moses and Solomon would be just
another game with its own rules.
However, before further criticism, a more extensive sample of
his views on logic and language is appropriate. The following
comes from paragraphs 89-109:
In what sense is logic something sublime? For these seemed to
pertain tologic .. . auniversal significance. . . . For logical inves-
tigation explores the nature of all things. .. . It takes its rise...
from an urge to understand the basis or essence of everything
empirical. .. . We donot seek to learn anything new by it. We want
to understand something that is already in plain view. [Isn’t the
understanding something new?]... We feel as if we had to pene-
trate phenomena... the possibilities of phenomena.... Our
investigation is therefore a grammatical one. . . . Misunderstand-
ings concerning the use of words, caused, among other things, by
certain analogies between the forms of expression in different
regions of language. ... But now it may come to look as if there
were something like a final analysis of our forms of language ...a
state of complete exactness. .. . When this is done the expression is
completely clarified and our problem solved. ... We eliminate
misunderstandings by making our expressions more exact. ... We
want to say that there can’t be any vagueness in logic... . But what
becomes of logic now? Its rigor seems to be giving way here. But in
that case doesn’t logic altogether disappear? ... We are talking
about the spatial and temporal phenomenon of language, not about
some non-spatial, non-temporal phantasm.... Philosophy is a
battle against the bewitchment of our intelligence by means of
language.

These quotations, with the omissions indicated, are not supposed

to misrepresent Wittgenstein’s position. Those who wish to check
every word in these twenty paragraphs may be puzzled, here and
there, as to whether Wittgenstein is seriously stating his own
view or occasionally taking the role of an opponent. However, it
seems in the main that logic is reduced to grammar and that
philosophical problems can be, not solved, but dissolved by
reducing the proportion of grammatical errors and unintentional
ambiguities.
To give a concrete example, an attempt will now be made to
show how Wittgenstein’s method solves or dissolves the mind-body
74 THE LATER WITTGENSTEIN

problem. This is surely a philosophical or metaphysical theme.

Socrates was able to face death with composure because he
thought that death was the separation of the soul from the body
and that the soul would continue living in a better world. The
Hebrew Scriptures, beginning with the early book of Job, mention
a future life, not so frequently as the New Testament, but they
mention it nonetheless. Christianity reunites the resurrected body
with the soul in the eschatological future, indicating also that the
soul enjoys the presence of God in the interim. Not to mention
Augustine, Descartes had two created substances, mind and body,
the relation between which he tried to expose. Later on—the story
is well known—there were materialists who denied the existence
of a mind or soul; and more recently the behaviorists defined
mind out of existence by using the word to designate a bodily
function. How, now, does Wittgenstein handle this matter? The
difficulty is that his remarks are disjointed and scattered. Nor is
their import any too clear in some instances. Misunderstanding is
very easy, and a reader or critic can only do his best.
Early paragraphs in the J/nvestigations scrutinize the notion
that some strictly mental activity accompanies speech and other
bodily functions. Consider paragraphs 34-38:
Suppose someone said, “I always do the same thing when I attend
to the shape [of an object]: my eye follows the outline and I feel. .. .”
And suppose this person to give someone else the ostensive defini-
tion ‘That is called a circle,’ pointing to a circular object and having
all these experiences—cannot the hearer still interpret the defini-
tion differently, even though he sees the other’s eyes following the
outline, and even though he feels what the other feels? That is to
say: this ‘interpretation’ may also consist in how he now makes use
of the word. . . . For neither the expression ‘to intend the definition
in such and such a way’ nor the expression ‘to interpret the defini-
tion in such and such a way’ stands for a process which accompanies
the giving and hearing of the definition.

Here we interrupt the series of quotations to point out a factor

that appears in several places. First, the ostensive definition
centers on a word. How is the word circle used? One must study
whether the emphasis on words is a magician’s trick to cause the
circle itself to disappear. Then, second, why should any explana-
tion of the use of words imply the nonoccurrence of a mental
process that accompanies the hearing of the definition, unless, of
THE LATER WITTGENSTEIN 75

course, the hearing itself is mental. But if the hearing is indeed

mental, and not just a physico-chemical oscillation of the ear
drum, behaviorism is excluded. Let us see then how Wittgenstein
proceeds:
35. There are, of course, what can be called ‘characteristic expe-
riences’ of pointing to (e.g.) the shape. For example, following the
outline with one’s fingers or with one’s eyes as one points. But this
does not happen in all cases. . . . Besides, even if something of the
sort did recur in all cases, it would still depend on the cir-
cumstances. ...
For the words ‘to point to the shape,’ ‘to mean the shape’ and so
on, are not used in the same way as these: ‘to point to this book’ (not
to that one)... . Only think how differently we learn the use of the
words ‘to point to this thing,’ ‘to point to that thing,’ and on the other
hand ‘to point to the colour, not the shape,’ ‘to mean the colour,’ and
so on.
Once again, the suggestions of this paragraph may possibly be
worked up into a good argument against ostensive definition. But
though it is difficult to see what Wittgenstein means by the last
half of the paragraph, it is rather clear that he does not mean it as
a repudiation of ostensive definition.
... Do you also know of an experience characteristic of pointing to
a piece ina game as a piece in a game? All the same one can say, “I
mean that this piece is called the ‘king,’ not this particular bit of
wood J am pointing to.” (Recognizing, wishing, remembering,
etc.).
36. And wedo here what we do ina host of similar cases: because
we cannot specify any one bodily action which we call pointing to
the shape (as opposed, for example, to the color), we say that a
spiritual (mental, intellectual) activity corresponds to these words.
Where our language suggests a body and there is none: there, we
should like to say, is a spirit.”
This paragraph certainly seems to teach a thoroughgoing
behaviorism. Thinking, or meaning, intending something, or
feeling, are actually bodily motions and as such are covered by
the laws of physics and chemistry. The illusion of a mental content
derives from these facts: the bodily motions are so various that no
one of them is present in every case; they are often hidden from
our view; therefore common opinion suspects something non-
physical, that is, mental; but in truth there is nothing but body
and its motions in space.
76 THE LATER WITTGENSTEIN

Throughout the ‘ordinary language’ philosophy there runs a

puzzling oscillation. These philosophers, as in the previous para-
graph, argue that ordinary language is mistaken when it talks
Platonic metaphysics or Christian theology; then they try to erect
a theory of behaviorism by an appeal to ordinary language. But
behaviorism is itself a metaphysics. Its language, moreover, is
not ordinary language, for no common language, when speaking
of thought, intention, or feeling, has in mind the physical motions
of bodily parts. Nor does it help the theory to reduce thought and
intention to the status of words, mere words, vibrations in the air,
which as such can refer to nothing.
Reporting his opponents, Wittgenstein, many paragraphs later,
writes:
303. “I can only believe that someone else is in pain, but lknow it if I
am.”— Yes: one can make the decision to say, “I believe he is in
pain” instead of “he is in pain.” But that is all.— What looks like an
explanation here, or like a statement about a mental process, is in
truth an exchange of one expression for another which, while we
are doing philosophy, seems the more appropriate one. Just try—in
a real case—to doubt someone else’s fear or pain.
304. “But you will surely admit that there is a difference between
pain-behavior accompanied by pain and pain-behavior without
any pain?”—Admit it? What greater difference could there be?—
“And yet you again and again reach the conclusion that the sensa-
tion isa nothing.”—Not at all. It is not asomething, but not a nothing
either! The conclusion was only that a nothing would serve just as
well as a something about which nothing could be said. We have
only rejected the grammar which tries to force itself on us here.
The paradox disappears only if we make a radical break with the
idea that language always functions in one way, always serves the
same purpose: to convey thoughts—which may be about houses,
pains, good and evil, or anything else you please.

Whether there is any language that is not intended to express

thoughts or mental states may be worth considering; but can the
question whether or not a mind, soul, intellect, a person, will
enjoy a heavenly life after physical death be dismissed as a mere
matter of grammar? Note further:
305. “But you surely cannot deny that, for example, in remember-
ing, an inner process takes place.”—What gives you the impression
that we want to deny anything? When one says, “Still an inner
THE LATER WITTGENSTEIN en

process does take place here”—one wants to go on: “After all you see
rte?

Does Wittgenstein fudge a little bit here? An inner process can

be muscular or physical. Only a physical process could be seen.
Should not the hypothetical objector have said, “Do you assert or
do you deny an invisible, intangible mental process?” If then
Wittgenstein replies, as he might, “I neither assert it or deny it,”
could we not conclude that he has evaded the problem? This
suspicion of evasion is soon strengthened.
307. “Are you not really a behaviorist in disguise? Aren’t you at
bottom really saying that everything except human [physical]
behavior is a fiction?”—If I do speak of a fiction, then it is a
grammatical fiction... .
383. We are not analyzing a phenomenon (e.g., thought) but a
concept (e.g., that of thinking), and therefore the use of a word. So
it may look as if what we were doing were Nominalism. Nominal-
ists make the mistake of interpreting all words as names, and so of
not really describing their use, but only, so to speak, giving a paper
draft on such a description.
Again, one must consider whether the identification of one of
nominalism’s faults quite absolves the author of the charge of
nominalism.
These quotations and the comments on them are supposed to
stimulate a doubt—a mental doubt, not a vibration in the air—that
Wittgenstein has dissolved the mind-body problem. A final selec-
tion will now be given. Attention is called to certain statements
that seem so obviously untrue that one is completely baffled as to
why Wittgenstein made them.
577. We say, “I am expecting him,” when we believe that he will
come, though his coming does not occupy our thoughts [ital. his; how
can we expect or believe without thinking?] (Here “I am expecting
him” would mean “I should be surprised if he didn’t come” and that
will not be called the description of a state of mind.) [Why not a
state of mind? Is it a motion in the larynx?] But we also say “I am
expecting him” when it is supposed to mean: “I am eagerly awaiting
him.”

The remainder of the paragraph ‘imagines’ that there might be

a language in which ‘expecting with joy’ is one verb and ‘expecting
with fear’ is a different verb. It is not necessary to imagine: the
78 THE LATER WITTGENSTEIN

Navajo language has several instances where several such verbs

replace a single one in English. But does this show that we can
expect a man’s coming without thinking about his coming?
Then Wittgenstein continues: “578. Ask yourself: What does it
mean to believe Goldbach’s theorem? What does this belief consist
in? Ina feeling of certainty as we state, hear, or think the theorem?
(That would not interest us.)” But does it not interest one who is
trying to solve the mind-body problem? The only person it does
not interest is one who has a feeling of certainty that he has solved
the problem, or one who thinks that the Saturday afternoon
football game is far more important.
To continue paragraph 578: “. . . how does the belief connect
with this proposition?... what are the consequences of this
belief? .. . ‘It makes me search for a proof of the proposition.’—
Very well; and now let us look and see what your searching really
consists in. Then we shall know what belief in the proposition
amounts to... .580. An ‘inner process’ stands in need of outward
criteria.”
f, now, Wittgenstein is talking about how you can know that I
believe a certain theorem, just that and nothing else, itis plausible
to say that an inner process needs an outward criterion. Further-
more, it is plausible that you could never know for sure that I
believe. I might be as deceitful as Catherine de Medici and your
opinion of my belief would be only a wild guess. But, note, all this
by-passes the mind-body problem. Wittgenstein’s remarks do not
rule out an inner, mental process; but the instigator of St.
Bartholomew’s massacre makes the inner process plausible. A
less gory example is Socrates’ standing motionless for twenty-four
hours (was it) thinking about we know not what.
“583... . Could someone have a feeling of ardent love or hope
for the space of one second—no matter what preceded or followed
this second? . . . the word hope refers to a phenomenon of human
life. (A smiling mouth smiles only in a human face.)”
Of course, no one could have a feeling of hope, or a belief in a
theorem, without something preceding or succeeding the given
second. But this does not show that the past and future contexts
are physical and behavioristic, rather than mental and spiritual.
In fact, upon reading much of this the first time, one might
THE LATER WITTGENSTEIN 79

conclude that Wittgenstein is not discussing behaviorism at all.

Unfortunately, he has explicitly connected behaviorism with this
discussion. In at least one place he seems to claim that he is not a
behaviorist: he is only a grammarian dealing with words. But if
so, his observations are not examples of a method to solve meta-
physical problems. The following paragraphs point out the diffi-
culty, not only of your knowing my mind, but also of my knowing
my mind. Strictly speaking, this is a matter, rather, of my
knowing what my mind will be tomorrow. But none of it per-
suades me that I have no mind. Vibrations of the larynx, wiggles
of the dendritic processes, are not persuasion.
The conclusion is that although Wittgenstein has made many
very interesting remarks and has proposed many intriguing
puzzles, he has done nothing to solve, or even dissolve meta-
physical and theological problems.
 ~, ae

‘ > wanna bee eee

7 i? 0 ote CQtieeas ee.
= * -
’
‘l) cconeuereee Oar eam’
' a as" %
we Seka Lianne ee.
Loney ia Sa ati
< 4
ete ar ost i ;
at > OF ay ' by) ai ia, tat
be}
nan po
ihe
~
vat 2 es = Tem
att
ae eee a aa ie
¥

‘m
he « a”

- a. e Ly : — ie = Pj ~“

’ 1.

7]

——

ss

an

——

i >

A a 7

Oh <n
 PART TWO:
RELIGIOUS THEORIES
 OWT TAR: 4s
GIN aOR
THE secular language theories, not surprisingly, provoked a
reaction in defense of religious language. It could not all be
unmitigated nonsense. However much the descriptions of Homer
and the terminology of Augustine, the works of Luther and
Calvin, not to mention the sacred books of Islam and Hinduism,
and all the poetry and hymnology of the ages—however much the
wording may have obscured the reality, a motivation cannot be
denied. What they said at such great length must mean something.
Apparently the first twentieth century author to reply at any
length to logical positivism’s language theory was Wilbur Marshall
Urban, whose 750 page volume appeared in 1939. Neither Urban
nor his theory was Christian. It was simply religious. Later there
came a flood of smaller books written by authors who professed
some form of Christianity loosely so-called. E. L. Mascall pub-
lished Words and Images in 1957; Langdon Gilkey published
Maker of Heaven and Earth in 1959; but this is not the place for a
bibliography. These later writers seem to have been stimulated
positively by Soren Kierkegaard and Kar] Barth more than they
were negatively by logical positivism. What needs to be pointed
out, however, is the fact that these later so-called Christian
theories do not show any knowledge of the early nineteenth
century language theory of Horace Bushnell, to which they are in
several respects similar.
The procedure will now be
(1) begin with Urban as more directly opposed to logical positi-
vism. Next will come

83
84 RELIGIOUS THEORIES

(2) a criticism of E. L. Mascall, who indeed follows Urban

chronologically, but who also argues for a theory of much
earlier origin. Then we shall consider
(3) Horace Bushnell, followed by
(4) a sampling of the large majority of contemporary liberal
theologians.
 Wilbur Marshali Urban

THIS very interesting author judiciously begins with the basic

empiricism of John Locke. Locke had been surprised to find that
he could not complete his EA’ssay Concerning Human Under-
standing without investigating the relationship between thought
and words. Berkeley learned from Locke and concluded that all
or most of the confusions in philosophy resulted from the use of
words apart from the ideas they should symbolize. “Draw the
curtain of words to behold the fairest tree of knowledge.” Urban
states the problem very clearly: if all words originate in sense
experience, then when they are carried over into the nonphysical,
the problem of their valid reference to nonsensible ideas is
immediately raised.! Or, further, “the naturalistic and ultimately
behavioristic view of language which has developed of necessity
from Darwinian premises, has brought with it askepticism ofthe
word, a distrust of language more fundamental than any hitherto
experienced. The naturalization of language makes of it, in the
last analysis, merely a method of adaptation to and control of
environment, and denies to it ab initio all fitness for apprehending
and expressing anything but the physical .. .” (p. 31). Or, more
pointedly, can advanced mathematics be expressed in words, or
is there a gap between the word and the world? If medieval
nominalism denied universals, the new nominalism denies indi-
viduals (p. 34). Then Urban asks the questions he intends to
answer: (1) How is language a bearer of meaning; (2) How is

\Urban, Language and Reality (Freeport, N.Y.: Books for Libraries Press,
reprinted, 1971), pp. 27, 28.

85
86 WILBUR MARSHALL URBAN

communication possible; (3) What is the relation of logic to lan-

guage; (4) How can language refer to things? This monograph
has no intention of summarizing his 750 page answers; a few of
his suggestive ideas are all that can be included here.
Urban’s material may be roughly divided into two parts. There
are his refutations, not only of logical positivism, but of John
Dewey also and a few others; and second, there are his own
constructive efforts. Some of the former is almost essential to an
understanding of the latter.
John Dewey held that language changed brute animals into
thinking and knowing animals by creating the realm of meaning.
Urban asks, was there not first a realm of meaning for the
expression of which language was created?
If language created meanings, then obviously things could
have no meaning prior to language. Urban is willing to acknowl-
edge that some meanings are created by language, but he is
especially concerned to show that there must be prelinguistic
meaning. One example of this is the fact that a wolf will refuse to
eat a piece of meat in which poison has been hidden. In some sense
the wolf senses a meaning. The bait means death. For the animal,
however, the meaning is not detachable from the sensory thing;
for man, on the other hand, it is.
Although human beings also sense these animal meanings, for
example when we begin to chew a bad nut and spit it out, these
are for us extrinsic meanings also. John Dewey may say, the
clouds mean rain. But they do not mean rain in the same way that
a bad taste causes us to spit it out. The clouds are a sign of a
different physical event. If the clouds meant rain in the animal
sense, they would mean the action of seeking shelter, as the bad
nut starts us spitting. But if the clouds really mean rain, there
must be a certain individuation of both the sign and the thing
indicated, which is not present in animal meaning. When the
clouds mean rain, they do not necessarily mean seeking shelter—
we may already be indoors looking at meteorological instruments.
Meaning as a cue or stimulus to action and meaning as a relation
between asign and the thing signified are two different meanings
of meaning.
Or, in other words, the behaviorist theory that a thing causes a
WILBUR MARSHALL URBAN 87

reaction fails to distinguish mechanical habits from interpreta-

tion of signs. If the sign is merely causal, there is no interpretation.
Urban wants to insist on this point. Russell had said that
“meaning is an observable property of observable entities.” Others
of the same opinion say “meaning is directly perceptible like color
and sound”; it is “an object of direct perception.” Though he
acknowledges a sense in which this is true, Urban takes it as
confusing. Russell’s theory assumes things are given; but if only
sensations are given, it is meaning that transforms sense data
into things. Things are “ideal constructions” (p. 105). Meaning is
not something that is perceived; it is understood.
When, however, Urban turns from criticizing Russell and the
positivists, and suggests something constructive, one must be
cautious. He wants sounds to become words bearing meaning by
their similarity to the things they designate. The word buzz is an
imitation of the sound it signifies. Or, the inherent meaning of the
sound ache turns it into a word. Such onomatopoeic words are the
first words of a language. Now, while the zz in buzz may sound
like the noise of sawing, what about the letter B? Why should it
not have been fuzz? Then, further, it is hard to see the similarity
between the sound ache and a pain. Later when he asserts that
ouatou, a word in some primitive language, sounds like and isa
symbol of a stream, and that ouatou-ou-ou for the same reason
means ocean, it is hard to follow him.
There are, of course, onomatopoeic words; and there are also
metaphors. Urban mentions the transference of the word kid, a
young goat, toa human child. In fact, he says that metaphor is the
primary law of speech construction. Does this not seem somewhat
of an exaggeration? The introduction of the word its into the
English language is hardly the result of metaphor. Nor can the
declining use of cases in Greek as it changed to Koine with the
increasing use of prepositions be so explained. Nor the virtual
extinction of the optative mood. However, Urban’s opposition to
behaviorism is well based:
In the words of C. S. Lewis, “speech is only that part of behavior
which is most significant of meanings and most useful for com-
munication.”. . . The inability of the behavioristic theory to explain
even animal meaning, if in the concept of meaning is included the
notion of understanding or interpretation of signs, would inevitably
88 WILBUR MARSHALL URBAN

bring with it the conclusion that a fortiori it is unable to explain

linguistic meaning. ... The causal conception of meaning [in]
reductive behaviorism equates both the meaning of the thing and
the meaning of the word with our way of reacting. . . . All mean-
ing... is sufficiently accounted for by causal relations. ... The
taste of a caterpillar or the sound of a bell are, to be sure, caused by
the stimuli, but wnless the notion of understanding, or interpreta-
tion of sign as sign be left out of the notion of meaning, the plausi-
bility [of behaviorism] vanishes. ... When we come to linguistic
meaning... it is precisely understanding that is the sine qua non
of such meaning. . . . Thesound does not become a linguistic fact at
all until it is detached from its purely causal context .. . and this
detachment and mobility are not functions of the physical envir-
onment (pp. 129-1381).

Then Urban strengthens his attack on behaviorism by a discus-

sion of intention.
All this may be and isa valuable refutation of behaviorism, but
Urban’s view that words represent, rather than are symbols for,
things, has its own difficulties. Previously mentioned was buzz
and ouatou. He will not have words to be arbitrary symbols. The
symbol is imitative and conjures up the picture of the thing itself.
Therefore picturesque language is more adequate than concep-
tual: for example, “Theory is gray but life is green’ would be less
adequately expressed in conceptual terms (pp. 147-148).
Quite the contrary, so it seems. Such metaphors have to be
puzzled out and put in literal language before their vagueness is
dissipated. They make for good poetry, but not for good under-
standing. When he says of analogical predication, “I am bringing
to light some aspect... which could not be determined or
expressed except by such a transfer” (p. 179), he robs analogy of
all meaning. Unless the analogy is based on a literal and univocal
similarity, there could be no analogy at all. Urban indeed on this
very page states clearly enough the views of those who oppose him
(we abuse words when we use them metaphorically, a case of
equivocal predication; analogical predication is ambiguous), but
if some people agree with Urban, others think the view he states
so well and rejects so sharply is the literal, nonequivocal truth.
Similarly unacceptable is his dictum that “all words have origi-
nally unquestionably a physical reference [and] words for rela-
tions are primarily spatial in character” (p. 185). The term originally
WILBUR MARSHALL URBAN 89

may push the question so far back into primitive society that no
one can produce evidence for or against the thesis. But if God
gave Adam language for the purpose of worship, at least the word
God did not have a physical reference. Nor is it evident how the
relationship of ‘more wise,’ or ‘more witty’ could ever have been
primarily spatial. Or, for that matter, what spatial relation can
be found in wncle or cousin? It would seem that sometimes Urban
makes very general assertions without sufficient justification.
Urban returns to these points many pages later. All words have
a physical origin and a physical reference. After using such
words, someone intuits a value and by metaphor applies the
physical word to a new reference. We do not first intuit an object
and then express it: the expression is a constitutive part of the
knowing. Language creates the world of cognitive meanings.
This paragraph (p. 345) is mainly concerned with values, words
such as generosity, nobility of character, and moral values in
general. Even the word morality, though he does not use this
word as one of his examples, must have had a physical origin,
which then creates a cognitive meaning.
This sounds implausible and indeed self-contradictory. It is
implausible because without the ‘intuition’ of an object, there
would be no stimulus to expression. Why or how could anyone
invent a word, other than a nonsense syllable, if he had nothing to
express? The cognitive meaning must come first and its symbol
second. As the science of electricity was being formulated in
early modern times, the experimenters noticed certain relation-
ships. To that date, no names had been given them. They had not
been known before, so that there was nothing to give names to.
But when the ‘intuitions’ occurred, the scientists took the names
of three of their own number, Volta, Ampere, and Ohm, and
assigned these names to the level of energy, the quantity of
current, and the resistance. Only after a person has a thought,
can he give it a name.
Not only is Urban’s theory thus implausible, but it is also
self-contradictory because he cannot avoid the difficulty that
made it implausible. “All words are physical in origin. . . . It is
through metaphysical transfer that they acquire their new refer-
ences ... they become the vehicle for the intuition and descrip-
tion or expression of new entities.” Here we have the intuition or
knowledge first and the word comes second. Buta page and a half
90) WILBUR MARSHALL URBAN

later Urban says, “any intuition of reality ... without an element

of description is pure myth. . . . Knowing in any significant sense
of the word is inseparable from language .. . language creates
(ital. his] the world of cognitive meanings. . . . Intuition is impos-
sible without expression . . . the expression is rather a constitu-
tive part of the intuition itself. ... One does not first possess an
object in knowing and then express the nature of that object in
terms of arbitrary and conventional signs, but the expression isa
constitutive part of the knowing itself” (p. 347).
Aside now from the fact or, if you wish, the appearance of
contradiction between pages 145 and 147, the latter seems false.
Urban uses Croce and aesthetics for support. “The artist does not
first intuit or present his object to himself and then find linguistic
or other forms with which to express it.” The present writer
deems this to be false because one of his hobbies is oil painting;
and he always selects his object first and then tries to find a form
by which to express it. But no doubt Urban would reply: that is
why you are not much of an artist. However, the present writer
also writes, and if he does not have the object, the knowledge, the
argument, before he puts it in words, any words he should write
would be much worse than those now found in his publications.
And if Urban had not thought before he wrote, I could not imagine
how he could have completed so interesting a book.
Without doubt Urban is indeed interesting; and a great deal of
what he says is excellent. He tries to alleviate his notion of spatial
reference and pictorial representation by saying that poetry
conveys a meaning not expressible in any logical picture. Poetry
is pictorial but not spatial. A map is a spatial picture, but
Turner’s painting of Venice is not spatial. Now, it would seem
that a spatial or pictorial theory needs considerable alleviation;
but Turner’s painting and all other landscapes as well are surely
spatial. They may alter the actual proportions of buildings or
trees; they may disturb perspective; but surely they are spatial
representations. What is better in Urban is his acknowledgment
that symbolic representations, chemical formulas, and musical
scores are not pictorial. Quite so; but then why extend the term
pictorial so far, rather than simply abandoning it as a theory of
spatial language?
More acceptable, indeed highly commendable, are some of his
remarks in refutation of nominalism. Scholastic nominalism, he
WILBUR MARSHALL URBAN 91

asserts, iS inconsistent because it still retained the reality of

individuals. Neo-nominalism abolishes all substantives: all is flux
and names distort reality. If universals are unreal, individuals
are too, for the mere naming of a thing is a minimal universal.
Neo-nominalism therefore has no things, but only ‘events.’ But
this makes nonsense of (1) perceptual meanings, (2) value mean-
ings, and (38) descriptions; then (4) because it makes nonsense of
metaphysics, it makes nonsense of all empirical meaning, for the
former conditions the latter (pp. 354-369).
Further, in opposition to his opponents Urban remarks that it
is an assumption of evolutionary naturalism, not a necessity of
logic, that language is purely practical. Even if it had originally
been such, it may have developed other uses since. But even at
first language was not purely practical: it was made for human
communication, and this is wider than mere practicality.
The difficulties in Urban’s extensive material seem to spring
from the opposition between his basic empiricism and its un-
wanted implications. Meaning and verifiability are inseparable,
he says; but observation is not the sole method of verification.
Mere sense data are not knowledge. No sentence is purely osten-
sive. Direct verification is a myth. It is one sentence that verifies
another sentence. Sentences are needed to interpret the sense
datum. For example, an observation of Mercury as verification
involves a host of presuppositions. The isolated observation cannot
verify. Therefore, also, perceptual truth is only probable.
Sometimes Urban is not only perceptive, but witty as well. The
several theories of truth, he says—correspondence, coherence,
and pragmatic—cannot be sustained by their own criteria. Cor-
respondence cannot be shown to correspond to truth. Coherence
coheres with nothing. And the pragmatic theory does not work.
Therefore the neo-positivists conclude that the meaning of truth
is a meaningless question, for if meaning is always reference toa
sensory object, truth can have no meaning because it refers to no
object: the truth of the criteria is truth only of interpretation
(p2501)
Along with wit and insight Urban’s constructive theory con-
tains much that is puzzling. It seems that originally language
with its onomatopoeic words was always poetry, or at least aes-
thetic expression. Science came later. Even though he so extends
the term poetry to include prose as well, its language is neither
9? WILBUR MARSHALL URBAN

cognitive nor practical. Its power is the power to evoke images.

The intellectualistic fallacy views the aesthetic symbol as an
imperfect substitute for philosophic or scientific knowledge. On
the contrary, Urban asserts, the symbol contains an unexpressed
reference which the abstract concept cannot express.
But is this so? Keat’s Ode to a Grecian Urn, though its abstract
concepts in the last two lines are utter nonsense, expresses some-
thing about a moment detached from the flux of time. It does not
express it very well, and we can hardly be sure what Keats had in
mind. Probably he was somewhat confused. But had he thought
clearly and expressed himself intelligibly, a reader, no matter
how poetical the poem, could have put the meaning into clear
conceptual prose. What cannot be expressed clearly is not mean-
ingful. The same is true of all art. Take the painting Angelus. In
our civilization the attitude of the two persons is recognized as the
attitude of prayer. Presumably, because they are French pea-
sants, the painting pictures Roman Catholic devotion. But the
painting does not convey this information. Present such pictures
to people who know nothing about France or Roman Catholicism,
for example a Tibetan monk or aJapanese shogun of last century,
and they can only ask, What does it mean? The abstract or con-
ceptual statement is far clearer than any picture can be. One of
Urban’s examples is Pascal’s phrase, “Man is a reed, but a
thinking reed.” Urban continues, “To say that man is a reed
is... biological(ly) grotesque. Yet it is by precisely such devia-
tions from the real that certain aspects of reality, otherwise
inexpressible, are actually expressed” (p. 474). Now, Pascal, the
mathematical genius, could, when he wished, use metaphorical
and poetic language. But to say that his meaning is “otherwise
inexpressible” is not biologically, but intellectually grotesque.
One can say, “man’s bones are more easily cracked than granite
and a drop of water properly placed can kill him; yet neither the
rock nor the water can think, and thinking is infinitely superior
to mere physical existence.” The aesthetic quality is here missing,
but the thought is nonetheless more intelligibly expressed; and
the thought is superior to mere aesthetic enjoyment. Urban also
quotes four lines from T. S. Eliot as “expressive or revelatory to
an extraordinary degree.” The first two lines are intelligible; the
third may be guessed at; but the fourth expresses or reveals
nothing but the unintelligible confusion of Eliot’s mind.
WILBUR MARSHALL URBAN 93

Repeating the idea “not otherwise expressible” four pages later,

and after another ten pages of fairly clear expression of his ideas,
Urban defends himself by saying, “In that symbolic form an
aspect of reality is given which cannot be adequately expressed
otherwise. It is not true that whatever is expressed symbolically
can be better expressed literally. For there 7s [ital. his] no literal
expression, but only another kind of symbol” (p. 500). To which
the intellectualist replies, “The cat is black.”
Anyone who then says, “The cat is black” is poetical metaphor,
does not deserve a literal denial.
For the purpose at hand, this monograph cites Urban as a
defender of religious language against the positivists’ assertion
that religion is meaningless. The reader may now anticipate how
he does so. Religion and poetry are closely related, but not identi-
cal. The emotion of religion has the quality of ‘the holy,’ which
poetry does not necessarily have. Religious language is not only
evocative, but invocative as well. Thus religion has a personal
God and is therefore dramatic and mythic. As the poet gets
something of reality which the scientist has missed, so the reli-
gious person gets something the poet has missed. Religious
language communicates something other language cannot.
These several of Urban’s phrases are at best ambiguous. Of
course, the poet gets something the physicist qua physicist has
missed. Does not biological language grasp something chemistry
has missed? The language of football gets and misses what the
language of international diplomacy misses and gets. The confu-
sion here is between language as such, and the various subject
matters of conversation. Of course, chemistry is not botany; but
the language of all of these is the same English language. Urban’s
statement therefore is true and trivial. What he really means, as
his own wording in one place indicates, is that religion is emo-
tional. Religion has no place for thinking. It is not intelligible.
God cannot be known. Belief has no place. This is surely not
trivial: it is simply false. If it is not false, then Christianity is nota
“religion.”
The religions Urban can identify all speak the same language
and are immediately friends. But Christ had religious enemies.
He said, “No man cometh to the Father, but by me.” Christ, of
course, was not very religious. Well, maybe Jesus was religious
94 WILBUR MARSHALL URBAN

once in a while, for “Holy Communion . . . is a simple piece of

symbolism to express a number of spiritual truths too great for
ordinary language. . . . Thesymbol expresses somethingtoo great
for words” (p. 586).
Now, the present writer, whose theology is known to a certain
public, may not and does not claim to understand all the logical
implications of the Lord’s Supper; but unless he had a literal
understanding of some of its intellectual meaning, he would have
no reason for going through the motions.
In opposition to sacramentarianism where the magic works
apart from understanding, the Scripture says, Let aman examine
himself—an intellectual task—for he that eateth and drinketh
unworthily eateth and drinketh damnation to himself, not dis-
cerning the Lord’s body—an intellectual task. Therefore the
Covenanters and Calvinists will not celebrate the Supper without
a sermon to be understood. As Calvin said, an implicit faith is no
faith at all.
More broadly, if Adam and Eve (p. 590) are merely mythologi-
cal or metaphorical expressions of man’s emotional alienation
from an unknowable God, then Jesus is simply a character in
Aesop’s fables to represent a fanciful union with that God. This
implication by itself does not refute Urban’s religion, but it shows
that it is not the Christian religion. The implication also suggests,
in fact Urban’s religious theory demands, that his religion be
irrational (and repulsive). Though he himself had said that God is
personal, this statement also must be mythological and has no
intelligible meaning. Its “concrete terms” are not what religion
“really says” (p. 621). Quoting Brightman with approval, Urban’s
“idea of God symbolizes [ital. his] a unity or harmony between
existence and value” (p. 624). But no doubt this impersonal
harmony is itself also a literally untrue myth. Urban indeed tries
to salvage religious language from the nonsense of logical positi-
vism, but he so empties it of all intelligible meaning that the
result is no better.
 2:
E. L. Mascall

IN the introductory paragraphs to Part Two, it was noted that

Kierkegaard and Barth had a compelling influence over most of
the later authors. The work of E. L. Mascall, however, will prevent
us from assuming that this influence was completely universal.
His only reference to Barth says, “It is not surprising that Dr.
Kar! Barth’s slogan, Finitum non capax infiniti, went together
with a denial not only of the possibility of natural theology (that
is, of any knowledge of God acquirable by man’s natural powers)
but also of any rational understanding of revelation” (Words and
Images, p. 104). We must now outline the matrix, that is, the
theory in which this quotation is embedded.
In Mascall’s book, Words and Images, the Foreword raises “the
question whether the utterances which we make when... as-
serting or denying that God exists have any significance what-
ever.” In anticipation of the Thomistic view he will defend, he
acknowledges that Christian philosophers “at their best . . . had
always been willing and indeed anxious to admit that there was
something very peculiar about theological assertions . . . andthe
medieval theologians had constructed a department of logic—the
doctrine of analogy—expressly to deal with this fact” (p. vil).
Mascall begins with a long critique of Ayer’s verification prin-
ciple. His fourth point is that Ayer uncritically limited “expe-
rience” to sensation. This limitation, however, cannot be a
generalization from experience because “mystical experience in
the broadest sense... contradicts it and [this] certainly ought
not to be dismissed without detailed examination. But in Ayer’s
book such an examination is nowhere made” (p. 10). Mascall

95
96 E. L. MASCALL

defends mystical experience. This was Mascall’s fourth point.

His other arguments are also important, though they are not
necessary to an exposition of his theory of language.
However much Mascall opposes the verification principle, and
the restriction of meaningfulness to tautologies (definitionally
true) and sensorily verifiable statements (meaningful but some-
times true, sometimes false), he insists on empiricism. Experience
is the sole source of knowledge, but experience is not always
sensory. Even “sense experience itself may consist of something
more than experience of sense-objects” (p. 31). Also “there may
be experience which is not expressible in sentences at all, or
which is expressible only in sentences of a very peculiar kind”
(p. 31). Locke, Berkeley, and Hume, in addition to or because of
their sensory definition of experience, denied to the intellect any
activity other than inference from sensation: “the intellect in no
way apprehends, it merely infers” (p. 33). Against this Mascall
asserts, quite apart from any mystical experience, that the
sensible “particular ... isnot the terminus of perception, not the
objectum quod ... but the objectum quo, through which the intel-
lect grasps, in a direct but mediate activity, the intelligible
extramental reality, which is the real thing” (p. 34).
It is interesting to note that Mascall goes beyond (as he says)
Thomas and refused to assume that “the real intelligible world
[is] isomorphic with the subjective sensible one” (p. 41). For
example, relativity and quantum theory are not statements about
sensible phenomena: they are “expressions of the kind of intelli-
gibility that the real world has.”
Then Mascall adds a short defense of mystical experience, the
language of which is entirely unintelligible to all who have not
had the experience. Hence it is useless to discuss it.
Any theory of language depends on a view concerning the
extent of possible knowledge and the methods of learning. Basic
to Mascall’s position is the assumption that “the intellect does not
only reason, but also apprehends; it has, as its objects, not only
truths, but things’ (p. 63). Understanding as ratio is the power of
discursive, logical thought, of drawing conclusions; understand-
ing as intellectus is a simplex intuitus, a simple vision to which
truth offers itself like a landscape to the eye. “Intellectus is not
E. L. MASCALL 97

concerned simply with the apprehension of a purely spiritual or

ideal realm; it is concerned equally with the perception of the
everyday world of material things” (p. 64).
In order that the reader may not confuse the position of the
present writer with those of Mascall, Gilkey, or others, it seems
allowable to interpolate into this exposition some preliminary
but basic criticism.
From the time of Parmenides down to the present, various
philosophers have based their epistemology on an analogy be-
tween eye-sight and knowledge. The object of knowledge, the
thing known, has been regarded as an individual object, some-
what like a tree or a rock. This object can be “seen” with various
degrees of clarity. Poor eye-sight or fog prevents seeing it clearly,
though the field of vision includes its complete outline with the
background around it. One sees the whole of it, if not the details.
Thus God looms up before us ina fog. Bonaventura described it as
a global representation of which the intuition is lacking. Mascall
presumably wants to preserve some sort of intuition.

But before considering such a difficult object as God, let us

apply Mascall’s view to a tree or a rock. He says the intellect
grasps material things: we can know the tree. Now if one should
challenge him to prove that he knows this individual tree, what
could he say but that it is an oak tree, about forty feet tall, with
leaves shaped like those of other oak trees, whose wood is coarse
grained and very hard? Unless he says something like this, would
we not conclude that he does not know this individual tree? Of
course, he could be less botanical and say simply, the trunk is
brown and the leaves are green.
Note that when he tells us what he knows, he gives us sentences.
For the sake of argument, we agree that the sentences are true.
But the words “oak tree” are not true. A noun all by itself is
neither true nor false. Knowledge (and is not knowledge the
possession of a truth?) always comes in propositions. Otherwise
language could not express a truth. Therefore the intellect does
not grasp individual material things. It is also impossible to know
mental “things,” if there are such. Is the concept of two an indi-
vidual thing? Whether or not, the concept of two, all by itself, is
unknowable. ‘One plus one equals two’ can be known, and we
98 E. L. MASCALL

assert it as atruth; but the number tivo, alone, like the oak tree, is
neither true nor false. The content of knowledge is always propo-
sitional. This view allows the intellect to do something else besides
drawing conclusions. It can know premises as well as conclusions.
Call it simplex intuitus, or contemplation, or understanding, it is
different from drawing out an implication. Axioms can never be
conclusions. But all truth comes in propositions. One can some-
what anticipate how this view of truth can apply to a theory of
language.
After some twenty or more pages of Thomism, Mascall is ready
to discuss “the relation of words to thought and communication”
(p. 88). First, he repudiates the notion that language is a code, in
the sense that one person puts his thoughts into a code and the
other person decodes them again into thoughts. We shall not
boggle at the necessity of both persons’ knowing the same code.
With all the ambiguities of English, “neither of them will apply
the name ‘tree’ to the object to which the other will apply the
name ‘tin-opener.’” The two persons learn the code “by a process
in which ostensive definition by our elders plays a large part.”
This code theory, Mascall continues, is not wholly false, but it
has “only a very limited validity” (p. 91). Its basic defect is the
same as that of the sensationalist theory: it confounds the objectum
quod with the objectum quo. This defect becomes clear when one
inquires as to how the code was originally set up. Ostensive
definitions are impossible unless the learner already knows that
ostensive definition is going on. Dogs don’t know this. Linguistic
formulae, like sensible particulars, are neither objecta quae of
communication (for then they would be mere flatus vocis) nor are
they merely more or less accurate structural replicas of thought;
rather they are objecta quibus: they are the means by which two
minds can share in acommon intellectual life. Cor ad cor loquitur.
Mascall as a learned Thomist must use the Latin code: English
says, Heart speaks to heart (p. 92). But neither English, nor
Latin, nor Masceall, is very clear on this point.
This vagueness allows him to say that the techniques, the codes,
of poetry, painting, science, and theology will all be different.
Each is adequate to its particular function. Nor is the function of
art merely to evoke emotion, and not to communicate truth. With
less vagueness, hopefully, the present writer agrees that the
E. L. MASCALL 99

function of art, even music, is to express truth; but maybe not just
as Mascall intends. Further, Mascall is not so much interested in
the languages of music and painting, as he is in theological lan-
guage, where the transcendent God stands out in sharp contrast
with every other subject of human thought and discourse (p. 93).
Mascall gives two examples of theological language. First is
the poetry, and prose commentary thereon, of St. John of the
Cross. In spite of the artificiality of the technique, every reader,
says Mascall, must be struck by the coherence and profundity of
the Carmelite Doctor. “What justifies a particular descriptive
technique is not its conformity to a predetermined criterion, but
its simple capacity to get its stuff across” (p. 95).
Some people, however, would say that St. John leaves them
completely baffled as to his meaning.
The second example is that of some “Protestant controversial-
ists” who have trouble with the epistle to the Hebrews. They
argue that Christ relinquished his office of priest because after
his ascension he sat down on the right hand of the heavenly
majesty. Priests stand; kings are seated. No one can be both
seated and standing at the same time. Therefore Hebrews is
inconsistent because it describes him as sitting and also as a
priest. Mascall, to defend the consistency of the epistle, argues
that analogies are not to be taken literally. Analogies literally
understood may conflict, but yet convey consistent meanings. To
some extent this is true, if analogies convey enough meaning. But
would it not be better to end this puzzle before it begins? Some-
times priests sit down and sometimes kings stand in their chariots
as they ride into battle.
Of course, it is true that analogies are not to be taken literally;
but here is where the theory of analogy must engage in mortal
combat, rise from its seat, and be laid low. Mascall himself states
the problem: how “a figure which is used only analogically
manages to describe its object at all” (p. 96).
To the critic Mascall’s answer seems to be an admission of
complete defeat. The Atonement, he says, is variously explained
by Christians and Christian churches. There really is no one
clear-cut doctrine of the Atonement. Several incompatible pic-
turesque images are in use, none of which applies univocally.
100 E. L. MASCALL

Each analogy applies only up to a point and no further. Yet this

does not imply “that theological discourse is insufferably impre-
cise” (p. 97). The figures of speech may awaken a man to the
realization of his alienation from God. Then when he enters into
the sacramental fellowship of the Church, he will experience the
Atonement. “The mystery will now be known obscurely and
imperfectly, it is true, but no longer imprecisely.” The Atonement
is then “known not by description but by acquaintance.”
The question was, How does a figure of speech, an analogy,
manage to describe its object at all? Does not Mascall reply, It
doesn’t? Since he insists that there is no orthodox doctrine of the
Atonement, since therefore several people mean different things
by the term, the analogies apply equally to incompatible doc-
trines. Or, better, they apply to nothing, for no one idea is
acknowledged as being pictured.
If the doctrine of the Atonement were clearly known, a preacher
might use a pleasing analogy or illustration that might attract
his congregation and help fix the meaning in their minds. But
suppose none of them have the least literal notion of what doctrine
X means. This might not be the case with some well instructed
congregations, but it was certainly true on many foreign mission
fields in the ninth or nineteenth century. Now, then, says the
missionary, I want to explain to you doctrine XY. None of them had
even heard the word X before. So the missionary says, X is like
the dawning of the morning. One of his audience thinks, X is an
event that happens approximately every twenty-four hours.
Another in the audience thinks, X is something reddish-orange.
A third guesses that X is a work of art, though not necessarily
reddish-orange. A fourth supposes that X is a method of locating
east. But since none of them have any knowledge of the literal
meaning of X, they have no way of determining in what respects
X is like the dawn and in what respects it is not. Analogies
require but do not furnish information.
Mascall is forced into his otiose theory of analogy by his basic
concept of the origin and use of language. “God is by definition an
infinite and suprasensible being, while all the language that we
have in which to talk about him has been devised in order to
describe and discuss the finite objects of our sense-experience”
(p. 101). But if one rejects this view of the origin and purpose of
E. L. MASCALL 101

language, if one maintains that language is a divine gift for the

purpose of conversing literally with God—as well as for counting
sheep—then he does not entangle himself in ineffective illustra-
tions in order to talk about God or even the number two.
A paragraph in the book almost admits this, but Mascall seems
to miss the import of his own words: “We must recognize that
thought about God—knowledge of God—precedes discourse about
him. If we could not ‘now [ital. his] anything about God, we
certainly could not say anything about him. And the possibility of
knowing God is intimately bound up with the doctrine of creation”
(p. 103). Mascall makes this excellent statement to prepare for
his rejection of Barth’s view. It is pertinent. But does it not also, if
slightly less obviously, dispose of Thomism as well?
Mascall then concludes by noting the undeniable fact that the
Bible uses images. It is not so undeniable that such images are
“objecta quibus in the cognitive process.” Further, the cognitive
process Mascall has in mind does not seem to be cognitive, for he
quotes Farrar with approval to the effect that if we seek theologi-
cal propositions, if we try to deduce from Scripture a logical
system of doctrine, “we close our ears to the voice of Scripture”
(p. 115). What we need is “the life of spiritual images.” In hisown
words Masceall continues, “for the understanding of images it is
not necessary for us to get behind them to a non-metaphorical
understanding of fact. The images themselves illuminate us”
(p. 116).
These statements are not meaningless nonsense as the positi-
vists would claim; they are just plain false.
 >
Horace Bushnell

SoME conservative theologians may still dimly remember Bushnell

as one who preached the ‘moral influence theory’ of the Atone-
ment; but few if any, conservative or liberal, ever heard of his
theory of language until Donald A. Crosby published Horace
Bushnell’s Theory of Language (Mouton: The Hague, 1975). Dis-
agreeing with that author’s evaluation, this study will nonetheless
profit by his scholarly investigations.
Language, for Bushnell, begins when some primitive man
attached sounds to physical objects. It was essentially a language
of nouns. Verbs begin as nouns denoting actions. All words origi-
nate in physical images. The word and came from an add. In
time—lengthy time is often used to solve problems that cannot be
explained—intellectual terms came into use. Physical objects
furnish the ground for the symbolism of intellectual discourse.
This development is not so successful as one might wish. Words
cannot properly represent even physical shapes. Words name
species, not individuals; in fact, they name only sensations. Hence
the inexactitude of the original physical language increases when
applied to the shapeless ideas of the mind.
Here an observation may be interpolated. Bushnell raised a
question as to how a code could originate. It is a legitimate
problem. But there is no difficulty in defending the adequacy of
codes. The letters d-o-g and the letters H-u-n-d, and the letters
c-h-i-e-n are all adequate to represent a certain type of animal.
Symbols are always adequate, just because they are symbols. It
seems useless to question the adequacy of theological language. If
theological thought can be defended, the language will take care

102
HORACE BUSHNELL 103

of itself. A person may indeed think of cat or God at the wrong

time; and he may say chien when he means chat, but this is no
defect in language as such. Therefore if one has an idea of the
shapeless number that solves the equation x? + 1 = 0, any symbol
will do.
However, for Bushnell, the shapes of physical objects are
supposed to represent, poorly, theological and mathematical
objects. Figurative language is clearer than literal. “No turn of
logical deduction can prove anything, by itself, not previously
known by inspection or insight” (Crosby, p. 28; Bushnell, God in
Christ, p. 58). Or, conversely, the logical arguments of the trini-
tarians are bad and irreligious; but the Unitarians are even worse
because their logic is better. Both should confine themselves to
images.
Bushnell seems to oscillate between language and the thought
it symbolizes, for he explains that logic—not the choice of
symbols—developed from grammar, and grammar came from
physical relations in nature.
This is a basic and fatal flaw in all empiricism. Even Aristotle
failed to give Aristotelian logic an acceptable basis. The reason is
that the laws of logic are universal. The syllogism Barbara is
always, everywhere, and without exception vaiid. But experience
is never universal. One may observe a thousand black crows, but
this is of no value in supporting the proposition, All crows are
black. The next crow may be an albino. Hence physical relations
in nature, if indeed they could produce grammar, would still
never arrive at any principle of logic, mathematics, or theology.
Maybe Bushnell had some vague inkling that this was so, for he
concludes that therefore language can apply to truth only in an
analogical sense. We need poetic insight. We come closer to the
truth only when it is offered “paradoxically.” Paradoxically is
Bushnell’s own term, some seventy-five years before Kar! Barth.
And it is much the same irrationalism. Poetry is better than
prose; the poet’s contradictions are all facets of the true. Inconsis-
tency is a positive good, for truth resides in feeling. The Gospel of
John is the best Gospel because it contains the greatest number of
contradictions. Creeds have some value for their own time, for
each one illuminates a given facet. We should, or can easily,
104 HORACE BUSHNELL

believe them all—the Assumption of Mary, the Swedenborgian

heaven, the pronouncements of Brigham Young, and the unver-
anderte Augsburg Confession.
When such criticisms are directed against Bushnell, a peculiar
situation arises. The author Crosby, in defense of Bushnell, com-
plains that the critics take Bushnell literally, whereas he should
be understood metaphorically. If he says language begins with
nouns attached to physical things, he does not mean that this is
true literally. If he says the laws of logic are abstracted from
limited experience, he does not really mean it. What the critic
should have done is to take Bushnell’s words “suggestively.” To
some people Bushnell’s words suggest nonsense.
When in opposition to Bushnell’s emphasis on feelings Hodge
says, “The whole healthful power of the things of God over the
feelings depends on their being true to the intellect. . . .The Bible
is not a cunningly devised fable—a work of fiction, addressed to
the imagination”; and when another author asserts that intellect
precedes sense and there is an intelligible world independent of
the human intellect; Crosby returns to defend imagination,
fiction, and feeling. “Certainty can only be had in an immediate
imaginative grasping of the truth” (p. 275, 276). Of course, Crosby
does not mean this literally. For him there are two ‘truths’; first,
ordinary information, but, second, truth is also a device to create
a response, and in this second kind of truth the question of truth
or falsity does not arise.
Crosby and Bushnell do indeed speak so as to create a response:
I feel, I have an imaginative grasp, I see intuitively that it is all
nonsense. This is a satisfactory reply because my feelings are
mine as much as his feelings are his.
 4

Langdon Gilkey

THE twentieth century religious language philosophers, as before

stated, seem not to have been influenced by Horace Bushnell.
Langdon Gilkey in Maker of Heaven and Earth (Garden City,
N.Y.: Doubleday, 1959; Anchor Books ed., 1965) makes no mention
of him. Yet one will observe several similarities.
The Vanderbilt and Chicago professor begins by branding the
early chapters of Genesis as fables and myths, reflecting “the
prescientific speculation of the Babylonian and Canaanite cul-
tures” (p. 27).! Then, to find religious value in Genesis, Gilkey
empties it and most of the Old Testament of intelligible thought.
This is accomplished by his constructing a theory of language
that pretends to preserve some religious significance in “all these
clearly paradoxical anthropomorphisms” (p. 320).
How paradoxical anthropomorphisms, and even Babylonian
mythology, can be religiously important without having any cog-
nitive content, Gilkey explains by saying that all language about
God is analogical. Or, to quote, “because it is inescapably analogi-
cal in character, theological language points to a meaning that
transcends any clear and precise description” (p. 67).
This explanation needs to be read carefully. It says, paradoxical
anthropomorphisms are useful in religion because all religious
language is analogical. But if analogical language has no cogni-

1The factuality of this assertion is convincingly contested by Oswald T. Allis,

The Old Testament (Nutley, N.J.: Presbyterian and Reformed, 1972, chapter V,
“Comparing the Incomparable,” pp. 341-378).

105
106 LANGDON GILKEY

tive content, how can it make anthropomorphisms comprehensi-

ble? How indeed could it make even nonanthropomorphic lan-
guage comprehensible? And if all religious language is analogi-
cal, what good is any of it?
Mascall, to whom Gilkey devotes a lengthy footnote on page
162, presumably held to St. Thomas’ theory of analogy.”
Bushnell and Gilkey apparently use the term vaguely to denote
any kind of similarity. These men do not seem to consider that the
statement of the similarity must be literal, not analogical; and
that without the literal basis no analogy is possible. What Gilkey
expressly says is that “theological language points to a meaning
that transcends any clear and precise description” (p. 67). This
notion of a pointer Gilkey probably borrowed from Emil Brunner,
or some other source in the so-called neo-orthodox or dialectical
school. But pointers are obfuscatory. First, in this connection the
verb point has no meaning. If Iput a mathematical problem on
the blackboard, I can point to it, or toa part of it, with my finger,
and explain the difficulty; yet pointing does not explain. But if
there is no blackboard, how can | point totriangularity, Rosaceae,
justice, or relativity? The opposition will reply that I am taking
the verb point too literally. But if I take it as a figure of speech
that means ‘imply, so that the relations among certain lines and
angles imply a theorem, the opposition will like it even less, for
this is an example of reasoning a outrance. What then do they
mean by it? Apparently nothing. Then second, how can they know
that the analogy points to anything? If I stick out my finger at
random toward the empty sky, I am not pointing at anything. To
know that one is pointing, one must see the object pointed at. But
Gilkey’s finger points to “a meaning that transcends any clear
and precise description.” So, the impossible pointer gives us a
nonmeaning. Gilkey may say that this is not his meaning. Maybe
it is not; but who can know, for his meaning or, better, his non-
meaning, is unintelligible.
Gilkey constantly emphasizes the unintelligibility of religion.
It isa“simultaneous affirmation and denial” (p. 335); “Whatever

“For a destructive analysis of that theory, ef. Gordon H. Clark, Thales to Dewey
(Boston: Houghton Mifflin, 1957), pp. 276ff.
LANGDON GILKEY 107

we say of him [God] must be affirmed and denied at the same

time.” God is both merciful and merciless; he is both omniscient
and ignorant; he has no spatial limitations and he is six feet tall.
“We cannot hope . . . todiscover how this likeness and unlikeness
are resolved; ... paradoxes are the only way of speaking about
God. . . . He eludes all our words and categories” (p. 336).
Perhaps then we were wrong in saying that God is merciful and
merciless, both spatial and nonspatial. We should have said that
God is neither merciful nor merciless, neither spatial nor non-
spatial, neither conscious nor unconscious. The name of the object
that fits this description is, Nothing. But Christians should not be
disturbed because “To attempt, therefore, to ‘smooth out the
paradox’ is fatal to the meaning of the Christian message”
(p. 340).

As the idea of a pointer presumably came from Brunner, and as

“smoothing out the paradox” is equivalent to Brunner’s rejection
of logic, a paragraph on this background will emphasize the
irrationalism of this school of theology. Paul K. Jewett (Emil
Brunner’s Concept of Revelation (London: James Clark and Co.,
1954, pp. 104-105) translates from Die Christliche Lehre von Gott
as follows: “The purely rational element of thought in logic has
the tendency to proceed from any given point in a straight line.
Faith, however, constantly bridles this straight line develop-
ment... . Theological thinking is a rational movement of thought,
the logical consequences of which are constantly, at every point,
through faith, turned back, curtailed, or destroyed. ... Only by
constant breaking of systematic unity and logical consistency .. .
does thought arise which may be designated as believing thought.”
Now, it might be polite to minimize, as hyperbole, the word
constantly which Brunner uses three times in this short quotation,
and also the phrase “at every point.” Surely Brunner allows a few
valid syllogisms; he cannot mean that faith permits no inference
at any point. Then substituting the words “every once in a while”
and “occasionally” the polite critic could give his respectful objec-
tions. But the present critic believes that Brunner really meant
what he said; and that the hyperbole cannot reduce “at every
point” to “every other point.”
How then can faith bridle, curtail, and perhaps destroy itsown
108 LANGDON GILKEY

logical consequences? Suppose we believe (for pistewo means

‘believe’) that Christ rose from the dead leaving the tomb as
empty as the women said; suppose further that this implies (as it
does when other scriptural statements are added in the premises)
we, too, shall rise. Is now our belief to curtail this inference and
even “turn back” and “destroy” the original premise? What sort of
“faith” would this be? If anything this “faith” or belief would be
unbelief. Yet this unbelief, this “constant breaking of systematic
unity and logical consistency” Brunner calls “believing thought.”
It is not believing; and if it is “thought” at all, the less the better.’
Although unintelligibility leaves no message at all, Gilkey, as
well as Brunner, will not retreat from this fundamental irration-
alism. Religious language is without exception mythological.
“Myth is a form of religious language which unites the three
concepts... of analogy, revelation, and paradox... . Thus when
Christians speak of God as Creator ...[or when they say] ‘he
sends his Son into the world,’ . . . he speaks through the proph-
ets. ..this mythical language is analogical because it . . . delib-
erately denies that the language is to be interpreted literally”
(pp. 342-3438). “Like the symbol of the fall, creation has no
inherent and original factual content” (p. 345). “If the ‘myth’ of
creation is taken to be literally and simply true, . . . then it loses
all its religious character” (p. 347).
On the next page Gilkey twice repeats his contention that if
creation is understood to be a literal fact, like an eclipse of the sun
in 1955, “it has no religious character. ... As an objective truth
about the world’s beginning, it has no deep reverberating bearing
on our own existence and destiny.” To this one may reply that a
literal creation has most certainly, if not an analogical and mean-
ingless deep reverberation, a most important and immediate
bearing on man’s life and destiny. It has as much bearing on our
future hopes and present conduct as the literal truth of a bodily
resurrection. But there is a more fundamental flaw, or at least
omission, in these pages. The quotations assume a knowledge of
what “religious characteristics” are. Where or how did Professor
Gilkey get his concept of the religious? If he can clearly state his

'Cf. Religion, Reason, and Revelation, pp. 146ff.

LANGDON GILKEY 109

concept, by what argument can he maintain it? He gives no

argument. Heis, of course, at liberty to invent any kind of religion
he likes. He is even at some liberty to misuse English and deny
that people of other religions are religious. He may compose, as
professors of psychology have sometimes done, a questionnaire
by which to test the religiosity of a hundred students. In one class
it was a Lutheran minister who was determined to be the least
religious—that is, who diverged the most widely from the profes-
sor’s notion of what religion should be. And this is all that Gilkey’s
language means. Orthodox Christians are not religious. Gilkey is
not Christian.
In fairness, for there is no profit in misrepresenting one’s oppo-
nent, one must note that in the last four pages of his book Gilkey
himself seems to have an incipient qualm. He asks, “If all our
knowledge of God is in terms of analogies, can we be said to
possess any significant knowledge of God at all?” This isa good, a
very good question; but the answer is pitiful. “The point where
God is most directly known,” he says, “is in historical revelation”
(p. 359). Naturally, for Gilkey, historical revelation is not the
historical event of God’s speaking intelligible sentences to Moses
or Isaiah. Nor can it be literal statements describing the Exodus
or the Babylonian captivity. He must mean uninterpreted occur-
rences. We cannot accept Moses’ explanation of the Exodus; we
can know only that there was an Exodus. Perhaps we do not know
that there was an Exodus, for there is so much inaccuracy and
mythology in the Old Testament. But let us suppose that somehow
we know that something, like an eclipse, happened. Now,
somehow or other, in such events as these—and in particular he
mentions the person of Jesus Christ, if only we could believe
anything the early Christians imagined about him—we see the
love of God. We learn that God is love—of course, he is also
hate—and love is not symbolic. “The personal recreative love of
God in Christ... is the one unsymbolic and direct idea of God
that Christians possess” (pp. 359-360).
Now, these seemingly beautiful words, as found in the matrix
of Gilkey’s book, are totally without meaning. The Apostle John
says, “Hereinis love . . . that he sent his Son to be the propitiation
for our sins.” But Son, sent, and especially propitiation are all
mythological. What Gilkey means by love, if anything at all, is
110 LANGDON GILKEY

something else. Now, if indeed ‘the love of God,’ in the matrix of

Gilkey’s theory, conveys any modicum of meaning, that meaning
cannot have been ascertained by any inspection of uninterpreted
historical occurrences. Consider two points.
First, observation of the historical process never produces any
ethical or theological ideas. World history, far from showing that
God is love, rather supports Bertrand Russell’s conclusion that
the world is unspeakably horrible. Not only did Hitler extermi-
nate five million Jews; Stalin starved ten million Ukrainians to
death; and the Chinese Communists massacred twenty or thirty
million Chinese, plus nearly all the population of Tibet. At this
point someone will ask, What about the person of Jesus Christ? A
good question; but the answer is not Gilkey’s. If we have no verbal
explanation of the person and work of Jesus Christ, he is but
another instance of the universal reign of injustice. The point is
that observation of the historical process does not come near
proving that the love of God is the one unsymbolic and direct idea
that Christians possess.
Therefore, second, this beautiful phrase has no meaning at all.
Besides the impossibility of deriving this idea from history, and
also besides the general rejection of the Scripture’s literal
message, there is another reason for saying that the love of God,
as Gilkey explains it, is meaningless. The reason is that the
alleged idea leads us nowhere. It points to nothing. Does the love
of God imply that as Jesus rose from the dead, so shall we? How
does one find out what faith, or repentance, or sanctification is,
from this vapid phrase? Does the phrase, ‘God is love,’ even tell us
anything about God? Of course, if we accept the literal and propo-
sitional revelation of the Bible, we can have an extensive theology;
but on Gilkey’s view we have neither a concept of love nor any
notion of why it is important. In the very same paragraph Gilkey
had also said, “In Christ God is not known as he is in himself.”
Hence nothing that Jesus is supposed to have done teaches us
anything about God.
Is not this sufficient to dispose of Gilkey’s mythical language
and irrational religion?
 5
Interlude

THERE are other authors whose theory of language is worked out

in somewhat greater detail than Gilkey’s. Before considering
them, however, it seems proper to sample the great number of
those who fall into the general category without originating
notable improvements in theory. The purpose is to take note of
types of evidence by which they support their views.
Many of these authors mention the first 11 chapters of Genesis,
and in particular the first three. When they say that chapter one
is not acosmological theory, they speak the truth, if by the phrase
they mean avery detailed description of the formation of the solar
system. This, however, is an irrelevance, just as much so as if one
were to say that the invasions of Sennacherib and Nebuchadnezzar
inII Kings are unhistorical because the book gives so little earlier
and later Assyrian and Babylonian history. The mere fact that
Genesis does not state the exact speed of light does not impugn the
statement that “he made the stars also.”
Another instance of poor logic relates to Cain. The author is
interested in finding inconsistencies in the Bible. So he argues
that Cain’s expressed fear of being killed by anyone who finds
him implies that there were inhabitants of the earth who were not
children of Adam and Eve. The gentleman should have taken an
elementary course in logic. His poor thinking ability is also
evident from another angle. If there were people not descendents
of Adam and Eve, especially if at all numerous, it is not likely that
everyone would want to slay him. Most of them would not have
known that Cain was a murderer. But Cain’s brothers knew.
Accordingly, by the biblical account everyone would want to kill

111
112 INTERLUDE

him because everyone had the same parents. Perhaps the modern
author thought that Cain had no brothers. He may have had none
at that time. Vengeance might be delayed; but it would come
because Adam “begat sons and daughters.”
The account of the Flood also stimulates a liberal to imagine a
major inconsistency. As ordinarily read, the Flood began on
2/17/600. It rained 24 hours a day for 40 days. After that, the
waters prevailed for 150 days. On 7/17/600 the ark rested. The
waters began to recede. On 10/1/600 the tops of the mountains
could be seen. Then Noah on three occasions let loose a bird. The
third time the bird did not return. Things were fairly dry by
1/1/601. And on 2/27/601, they all came out of the ark. So the
account reads. But somewhere between the lines this imaginative
author discovers an alternate version that limits the Flood to 61
days.
This sort of thing, if it were true, could legitimately be used as
evidence for Wellhausen’s documentary theory; but it is totally
irrelevant to the thesis that Genesis and all religious literature is
written in mythological language. The alleged contradictions in
the Bible are more clearly contradictory if the language is literal.
Then one can plainly say, this statement is false, or even that,
when the two are simply different rather than contradictory,
both are false. But inconsistencies never will prove that the
language is mythological.
What then can prove that this or that book is mythology? We
call Homer mythological because of a combination of two reasons:
he talks about the gods and we do not believe what he says about
them. Were we polytheistic believers we would not dismiss his
stories as myths; and when he writes on the Trojan War we may
doubt his accuracy, but we acknowledge that there was a Trojan
War. Now, it is possible to treat the Bible this way. Those who call
the early chapters of Genesis mythological do so because they do
not believe what Genesis says about God and his actions. To say a
religious book is myth simply means, I don’t believe it. In this
sense the books of Samuel, Kings, and Chronicles can also be
called mythological. But heresome embarrassment arises. These
books are, or claim to be, history, and they are as literal as any
history books are. From this it follows that these books are both
literal and mythological; and this is not what the language theo-
INTERLUDE GIS

logians want. For them mythology is something like a fable. It is

an analogy of an inspirational character that produces a moral or
otherwise desirable reaction. Now, the books of Samuel and Kings
are clearly not fables like Aesop’s; they are history written in
literal language. Because literal, and because one believes them
true, they are as inspirational as, indeed more stimulating than, a
merely fanciful tale.
The numerous unnamed authors referred to in this Interlude
differ among themselves in several details. Each one can be
evaluated only on the basis of his own words. But a survey shows
that their range of difference stretches from the easily understood
assertion that the Bible is literal language and often false to the
other extreme that the Bible is mythological throughout and
always true, ina mythological sense of true. Those who represent
this latter extreme face two problems: first, their own religious
writings are not literally true, and this makes their arguments
unintelligible; and second, since history can be religious, and
since mathematics was a religious and soteric activity for the
Pythagoreans, these authors have a hard time finding anything
literal in English, French, or symbolic logic.
Those in the middle, who take Genesis as myth but Kings as
literal, face the problem of stating a criterion by which to
maintain this distinction. If they say religious language alone is
mythological, they imply that Kings is not a religious book. Asa
matter of fact, they should also say that the first chapter of
Genesis is not religious, for cosmology is as irreligious as
history.
A particular example, slightly different in subject matter
though similar in its logic, may clarify the difficulty. In the
recent past several writers have said that the purpose of the Bible
is to present salvation in Christ. But since Kings and Chronicles
donot clearly do so, these books are not the Word of God. No doubt
most of John’s Gospel is the Word of God, but very little of Chroni-
cles. It is right here that the pointed question must be put. What
criterion is used to distinguish religious literature from nonreli-
gious? What criterion is used to determine that the purpose of the
Bible precludes historical books from being the Word of God?
Most of the authors who make these distinctions offer no criterion
at all. If they did, a Moslem or Hindu would reply, “that may be
114 INTERLUDE

your idea of religion, but it is not mine”; and a Christian would

reply, “that is your notion of what is excluded from salvation in
Christ, but it is not mine.”
There is indeed a way for these people to avoid logical difficul-
ties, paradox, and analogy. To quote one of them: “It is possible to
lead a religious life without discussing it or verbalizing very
much about it.” If a person never says anything, he obviously does
not flounder in fallacious implications. No one can refute him, for
he says nothing to refute. What one can truly say of him, however,
is that he is not a Christian, for Christ commanded his followers
to make disciples, “Teaching them to observe all things what-
soever that I have commanded you” (Matt. 28:20). Christians
must “verbalize” (to use contemporary gobbledygook).
 6
William Hordern

Speaking of God (New York: Macmillan, 1964) is a most inter-

esting attempt to expose the difficulties of some modern theories
of language. Like his predecessors Hordern is more successful,
very successful, in refuting antireligious views than in establish-
ing a solid base for Christian truth. The Christian can, however,
learn much both from his success and from his failure.
Hordern notes that there isa contemporary stress on communi-
cation, both in advertising and politics, but with this there is also
a debasement of the language in that words have become so
elastic that no one understands anyone else. Everybody favors
peace, freedom, and democracy (including the Democratic Repub-
lic of East Germany), but the words are used in contradictory
senses. Someone said that poor people are denied psychiatric
care, not because of their poverty, but because the psychiatrist
cannot understand what they say. One of the main causes of the
debasement of language is the sin of wanting to control the actions
of other people, which can most easily be done by propaganda.
Now, Christianity has a message to be understood; hence
Christians must be concerned about language. The question is,
How can we use words to speak about God? If we have coined
words to speak of earthly things, how can we use those words to
apply to heavenly things? The creeds have an historic strength as
a congregation repeats them; but the terminology is an offense to
modern man. The language alienates him. Candidates for the
ministry have a hard time subscribing to the denomination’s
creed—if they still have any scruples against perjury—yet they
want to take part in what they consider to be in harmony with the

115
116 WILLIAM HORDERN

denomination’s real purpose. So they pledge their vows with their

fingers crossed.
The fundamentalists resisted this claim to preserve what
somebody supposed was the real purpose of the church, to retain
what is essential and discard miracles for modern science. The
liberals won, but only to be eclipsed by neo-orthodoxy and its
myths. But if all is myth and symbol, where is truth?
If people in general found it difficult to discover what the
liberals really believed, it is harder to find out what the neo-
orthodox believe. They say they believe in ‘creation,’ but that this
tells us nothing about the universe. Words, of course, are symbols,
but this is not what the neo-orthodox mean when they say the fall
of man is a symbol. Ordinary symbolism and figures of speech
can be put into literal language, but the neo-orthodox doctrines
cannot, and these theologians cannot tell us what the symbols
symbolize. For example, Tillich said that the only nonsymbolic
statement that one can make about God is ‘God is being itself.’
Then someone replied, (1) this, too, is symbolic; (2) being itself is
not the Christian God; (3) the phrase is not a statement, but a
definition and one that is at odds with 95 percent of religious
tradition. Tillich then amended his statement about God to “the
only non-symbolic statement about God is that everything we say
about God is symbolic.”
Theologians have perhaps been dilatory in recognizing lan-
guage analysis. Formerly Christians faced the charge that their
theology was false or untrue; now they must face the charge that
it is meaningless. On the other hand, the analysts, who claim to
examine language as it is commonly used, ought to be willing to
consider ordinary theological language. For the most part they
slight it or ignore it. Those who ignored it, in that they dismissed
it with disparaging epithets, placed themselves in a peculiar
position. Their epithets are emotional, certainly they are evalua-
tive; but on their own showing such language is meaningless—
they were talking nonsense. In addition their verification princi-
ple was itself metaphysical and cannot be verified; yet from this
nonverifiable principle they deduced their system.
After some stimulating analysis of the analytical philosophy
Hordern begins to prepare for the exposition of his own position.
WILLIAM HORDERN ILA

He does not minimize, rather he very clearly points out the diffi-
culties in defending theological language. Predicates such as
angry, or verbs like know and will have strange meanings when
God is the subject. If we say God made or created the world, the
meaning is not the same as a statement that Smith made a chair
or table; for we can point in distinction to something Smith did
make, but to nothing God did not make. In fact, we cannot point to
anything God made all by himself, for everything now has been
modified by natural processes. Our knowledge, says Hordern, is
temporally conditioned, and for this reason does not apply to the
eternal God.
If theological language is not pure nonsense, one must give a
clear answer to the question, How can theological language
communicate meaning? The secular philosophers who have
attempted to answer usually have little understanding of theology
and as a result they present a caricature. The theologians, on the
other hand, know very little philosophy and as a result they are
incompetent.
Hordern’s key to the solution is the concept of conviction as
opposed to emotion. Conviction describes the state of mind of a
religious person better than emotion does. The theologian is con-
vinced. But, further, conviction presupposes a convictor, an irre-
sistible power outside oneself. Convictional language points to a
reality as much as empirical language does. Even in science there
is conviction about objectivity, and also conviction that science is
good. The logical positivist expressed his convictions by selecting
his verification principle. He was convinced that nothing was
important beyond the space-time world.
Note how this convictional theory leaves the logical positivist
breathless. He had argued that God and providence are meaning-
less because no experiment can invalidate them. When evil is
uSed as an argument against God, the theist makes some
unprovable assertions that this evil eventually produces greater
good. But though the positivist scorns the theist on this account,
he himself uses the same argument, for ne argues: Nothing can
falsify the scientific principle of order in the universe; for whena
difficulty arises (such as the perturbations of the moon or the
fickleness of the weather), the scientist claims that the present
experiments are defective in one way or another, there was an
118 WILLIAM HORDERN

error somewhere, and then he looks for new experiments until he

finds one he thinks fits.
The later Wittgenstein had spoken of language games. Unfor-
tunately he did not define game, and he seems to have used the
word in several senses. But at any rate, if the analytic philosophy
makes use of some concept of games, it cannot, as Hordern
emphasizes, rule out the theological language game. Baseball has
its rules; but this does not make football impossible. As there are
different rules for different games, so, too, theology establishes
its own rules—science cannot establish them for it. Yet theology
can prescribe some rules to science, for the latter, as was shown
earlier, cannot claim any value for itself. The language of values
is theological language. For this reason, though the theory of
language games embarrasses the positivists, it does not fit theol-
ogy. Theology not only evaluates science, it gives direction to all of
life. It does not answer questions left over by science, nor does it
offer an explanation of the universe; rather it isa means by which
aman can give purpose and direction to his life.
Along with conviction and convictor, Hordern will build his
theory of language on its use inacommunity. Analytic philosophy
usually ignores the persons who speak and their community. But
even scientific language becomes nonsense outside the scientific
community. The Christian community is the Church. Theology
cannot be understood if its Church background is ignored. Of
course, the gospel must be preached to those outside the Church.
Some Christians wish to lay a foundation for this preaching in the
evidence of natural theology. Then when the evidence becomes
weak or exhausted, faith takes over. Faith is a sign of weakness.
But, remarks Hordern, in the New Testament, faith is something
strong and powerful. It is neither emotive nor subjective. It is not
a rational act of choosing the most probable hypothesis. On the
contrary, “Here I stand!” This is grace, not the “Will to Believe.”
Wittgenstein himself had admitted that the giving of reasons
must come to an end somewhere. If a person is unwilling to play
the [scientific] game, no further reasoning can force him to. On
such grounds as these Hordern can infer that faith is not belief on
insufficient evidence: it is belief without any evidence. How does
one decide what evidence is? What evidence can one have to prove
his evidence is evidence? The very acceptance of evidence is faith.
WILLIAM HORDERN 119

Ayer admits the same thing in acknowledging that we cannot

logically justify the principle of induction, nor that the future will
resemble the past. Induction itself, says Ayer, sets the standard of
rationality.

Hordern’s argument in this section of his book is not altogether

clear. Or it is not altogether complete. He seems to say that one
man’s faith is as good as another’s. This, to be sure, is enough to
puncture the arrogance of the positivists. But the present writer
would desire to go further and insist, on the basis of the thousand
and first crow, that faith in induction is a very poor type of faith.
But to embarrass the positivists it is enough to say, as Hordern
does, that Western society has advanced in science because most
people believe it to be good and important. But the Hindus havea
different concept of reality, even though they presumably have
similar sense organs. Therefore the Hindus will sleep on a bed of
spikes to show that pain is unimportant.
Returning again to his constructive theory, Hordern reiterates
that theological language points to a convictor. This convictor is
known to be a mystery. When a Christian theologian speaks of
God as transcendent, he uses the word as asymbol of the mystery
of God. Resembling somewhat Schleiermacher’s feeling of abso-
lute dependence, this experience of mystery is the root of religion.
A mystery is not a riddle to be solved, says Hordern, it is asecret.
But even when revealed, it still does not become “transparent” to
men. One might say that a mystery is unknowable, or, better, that
mystery is not a matter of knowledge at all.
Such a defense of religion against scientism is popular in this
last half of the twentieth century. But though a defense of ‘reli-
gion,’ it is of little value to Christianity. For Christianity a mystery
is indeed a secret, and when God tells us the secret we know it.
“Behold I teil youa mystery,” Paul writes, “we shall not all sleep.”
No doubt most of the world’s population believes that all must die;
but the Christian on the basis of God’s revelation knows that at
Christ’s return some will not die. This is a known proposition, a
known truth—and easily understood. Those who deny it must
understand in order to deny. Now, indeed this known truth may
stimulate worship or awe; but it itself is an object of knowledge.
By removing God’s secrets from the sphere of knowledge, does
120 WILLIAM HORDERN

not Hordern produce a religion with which the logical positivists

have no quarrel? Would not these scientists allow anyone to have
any feelings he pleases? The difference between the logical posi-
tivists and Hordern is not with respect to knowledge, but with
respect to the evaluation of feelings. In awe Hordern asks, Why
should there be any world at all? Obviously (though it is not so
obvious) anything in the universe might not have existed. Science
can tell us why things are as they are today, given their antece-
dents, but it cannot answer the basic question why there is
anything at all. The person who tries to brush this question aside,
Hordern insists, is not unintelligent: he lacks a reverence for
reality. The analytic philosophy is totally devoid of mystery or
worship. It restricts language to what can be said clearly. But
clear speech and the cosmological argument are not religious.
Romans 1:19-20 has nothing to do with the cosmological argu-
ment: it refers to awe. “The sense of mystery is no proof of God. It
is compatible with many views of God, and even the atheist may
be aware of the mystery of the universe. .. . But it [the sense of
mystery] helps us to understand the use of theological language”
(p. 120).
At this point one may stand in awe of a mystery that supports
Christianity and atheism equally well. If this aids us in under-
standing the use of theological language, have we not already
accepted logical positivism?
Clearly Hordern thinks he is refuting his opponents. Actually
his position is very much the same thing. The theologian must say
the unsayable. In poetry, art, and liturgy we can sing what we
cannot say. The Te Deum is not paying metaphysical compliments
to Deity. Language is conditioned by space and time; the words
fail because they are finite. We say what we do not mean. We must
use analogies and then erode them. Yet ‘fatherhood’ is a better
analogy than ‘cousinhood’; but we cannot find the line that divides
the misleading part of the analogy from the good part. Theology
is paradoxical. College students reject theological language, but
they are moved by the Lord’s Supper (pp. 120-128).
But, then, college students are often irrational.
Naturally Hordern wou!d not look with favor upon the objec-
tions here levelled against him. He tries to avoid them by The
WILLIAM HORDERN Wail

Personal Language Game (Chapter VIII). But some of it seems to

contradict what he has already asserted.

Dogmatic theology and skepticism are both bad. There is a

base of knowledge from which faith can speak (knowledge, he
says). Skeptics say that until one has nonanalogical knowledge,
one cannot even know that something is an analogy. But there is
knowledge and there are analogies, he asserts with emphasis,
evading the sceptic’s, and as well the dogmatician’s, point.
Equally without support he asserts that there is a ‘personal
language game’ with its own logic, a game that points directly to
God. God has revealed that personal language is the key to
speaking about him. Science cannot speak of persons. Mind is
only one part of a person. We want to know not only how someone’s
mind works, but what are his hopes, motivations, and ambitions.
For personal language we must listen to the existentialists, rather
than to the language analysts. But if we listen to the existential-
ists, it is not likely that we shall hear anything. Do they explain
this personal logic, so different from Aristotelian logic? Does
personal logic have 24 valid syllogisms, or none at all? Does it
have middle terms, contraries, and subalterns? An assertion that
personal language has its own logic requires an exposition of its
rules. Hordern speaks of itas a game. Games have rules. Or is this
personal language nothing else than a different vocabulary?
Astronomy uses the words star and planet; botany speaks about
internal phloem. Brown is not a personal word, but sympathetic
and angry are. Then one asks, which verbs are personal and
which impersonal? A clock runs; and a boy runs, too. Without
explaining any of this, Hordern says, “The normal subject-object
division that occurs in our language about things is absent when
we are speaking about ourselves, or, if used, must be recognized
as having a different logic” (p. 137). To invent an example, ‘John
is. sympathetic’ contains neither a subject nor a predicate; and
because personal, it implies by its own logic that Scotchmen are
stingy.
There are some 50 or 60 pages more on which Hordern extols
personal language. Some of his remarks, especially in opposition
to Tillich, are valuable. But their matrix isa language game that
has no rules. Why should anyone want to play such a game? Chess
is infinitely better.
 7
Kenneth Hamilton

A recent book is Words and the WORD (Grand Rapids: Eerdmans,

1971). Professor Hamilton’s work is not so detailed as Hordern’s
nor so blatantly irrational as Gilkey’s. He develops his view care-
fully, step by step; but we may conclude that the results are about
the same.
His basic view of language, not merely religious language, but
all language, is stated clearly: “All language grows out of mythic
thinking and still bears the marks of its origin” (p. 86). This
thesis is repeated with some frequency and these other passages
fill out his thought. Quoting Wheelwright with approval, he says,
“Myth, then, is not in the first instance a fiction imposed on one’s
already given world, but is a way of apprehending that world.
Genuine myth is a matter of perspective first, invention second”
(p. 46).
Doubtless there is some truth to this; but it should also be noted
that the myth-makers did not recognize themselves as myth-
makers. They believed they were speaking the truth. When a
man throws his newborn son into the fiery arms of Moloch, he
must have very strong convictions. Only later ages can call the
stories myths, for only they believe that the myths are false. If
now Hamilton or someone wishes to ¢all all religion mythical, he
must mean that it is false. Let us trace it through his exposition
and see.
To bring myth into the present century Hamilton writes, “Each
life reenacts in part the history of the human race. Children
experience to some degree the formation of a meaningful world
through the mystic power of language” (p. 47). Here the expres-

122
KENNETH HAMILTON 123

sions “in part” and “to some degree” are extremely vague. It may
be that Wordsworth as a child invented myths for his later odes,
but the present writer does not remember anything remotely
approaching myth-making. Mythology was not at all “a marked
characteristic” of my childhood. Nor was it ever evident among
the boys I played with. But however this may be, and however
Homer and the Babylonians mythologized, it does not follow that
all language grows out of mystic thinking. Nor does anything else
prove that all words still bear the marks of a mythological origin.
If all language bears the marks of mythological origin and if the
words cat, triangle, baseball, and jetplane are a part of language,
they must show marks of mythology. A theory that makes such an
implausible assertion ought to be particularly careful to point out
these marks. Hamilton does not do so.
Hamilton seems to adopt Ernst Cassirer’s view that “Intelli-
gence... is not man’s decisive characteristic. What really dis-
tinguishes him from other animals is his ability to construct
symbols. ... He does not first understand the world, and then
learn how to put his knowledge into words. Rather his invention
of verbal symbols provides the possibility of his having knowl-
edge” (p. 45). But is not this patently backwards? It takes intelli-
gence to construct symbols; only because man is distinguished
from “other” animals can he do so. Moreover, before a man con-
structs asymbol, he must have something in mind to symbolize. A
primitive man would never invent the sound or symbol cat, unless
he had first seen a twitching tail and heard its other end say
‘meow.’ Does anyone believe that the savage said to himself, Catis
such a nice sound that I shall use it to symbolize whatever crosses
my path tomorrow at noon”?
Hamilton emphasizes mythology because his theory of religion
requires it. However, religious language today is not strictly
mythological. Hamilton proposes two advances in the history of
language before it can serve the purposes of present day, and
even some ancient, religion. The first step is to dilute or refine
myth into poetry. This advance gives us a God who really exists,
as opposed to mythological gods who do not.!

1This sentence is not given as Hamilton’s own words. It isa brief interpretation
124 KENNETH HAMILTON

Without paying too much attention to Hamilton’s view of poetry,

we do better to study what he considers truly religious language.
It is a step beyond poetry; viz., parable. “Christian faith...
gladly admits . . . the literal acceptance of myth untenable,” yet
man by reason of symbolic language remains “a myth-making
creature.” Then he continues, Christian faith gives “no privileged
instruction about ‘what the case is’ in the created world” (p. 67).
The ‘case’ of God’s command to Abraham to sacrifice Isaac is thus
excluded from knowledge. Admittedly the creation of the world
is excluded. The role of so-called Christian faith is for Hamilton
limited to “essential knowledge about the world as divinely
created. It also gives him assurance of the human meaning of his
existence. It mediates this meaning beyond the reaches of hisown
consciousness. .. .”
But if literal language is ruled out, no one can know that
literally God created the world. “Assurance of the human
meaning of his existence” would satisfy no man. At least a
thoughtful man would want to know what the meaning was. And
how a contentless faith can “mediate” an unknown meaning
beyond the reaches of consciousness is a problem for the psychia-
trist. No wonder Hamilton eschews plain, literal, intelligible
language in his religion.
Parables are supposed to solve this problem. Note, however,
that a theory of language, not simply a hermeneutic principle for
biblical exegesis, ought to show how mythological language devel-
oped into poetical language. After all, in Homer the two languages
are the same language; and if poetry gives us an existing God,
why can’t we find Zeus on Mount Olympus? Such a theory of
language ought also to show how poetic language develops into
parabolic language. Hamilton does not do this. Cannot parables
as well as poetry be mythological? It is not that there are different
languages: There are different subjects of conversation, expressed
in different literary styles, all in one language. This confusion
between subject-matter and style seems to be the result of a
decision to reject literal truth in religion. Hamilton is interested
in preparing the ground for an attack against plenary and verbal

of pp. 61-62, 67ff. Any different interpretation would be hard put to explain how
poetic language is intermediate between mythology and Hamilton’s religion.
KENNETH HAMILTON 125

inspiration. “ ‘Dictation’ theories of revelation sometimes seem to

assume that God communicates his Word through vocables, so
that understanding the exact sense of an aggregate of propositions
is to receive the Word of God. This is surely to bind the divine
Word to the measure of human words . . .” (p. 75).

This is a sentence the propaganda devices of which require

more than a paragraph to sort out. What he says about “dictation
theories” is probably not just “sometimes” true, but rather always.
However, it is also true of some theories that are not “dictation
theories.” B. B. Warfield argued against the modern concept of
dictation as practiced in a business office, but he held firmly to
the point that God communicates his Word through “vocables.”2
Then second, the idea that if God speaks in words, he is somehow
“bound” by an alien force, is a thoroughgoing misrepresentation.
The words themselves are mere signs or symbols. They designate
ideas or truths. If God cannot use symbols to express his truth, he
is indeed bound and limited. A God who cannot speak is not
omnipotent. In fact in such a case God would be more limited than
man, for a man can speak. Then, further, if God cannot speak, he
cannot speak parables.
It is quite clear that Hamilton does not accept the Bible as the
Word of God. “The fact that words are in the Bible. . . does not
mean that our reading of them necessarily must yield authorita-
tive statements that we can proceed forthwith to identify with the
Word of God” (p. 76). Well, of course, not necessarily, for some
people some ofthe time do not understand the words they read; so
that “our reading” the words, if we are such people, does not
necessarily yield correct propositions. The phraseology here is
again propaganda, for the important question is not whether
some people misread the Bible, but whether the words and sen-
tences of the Bible are authoritative statements because they are
true, because they are the words of God. It is obviously poor
thinking to attack a theory of the inspiration and truth of the
Scriptures on the ground that some people do not understand the
words. Must one take a textbook on calculus as mythological,

2On the liberal caricature of Calvinistic theology, cf. Religion, Reason, and
Revelation, pp. 115-119.
126 KENNETH HAMILTON

poetic, or parabolic and not literally true, because some high

school students cannot understand it? It is by such invalid reason-
ing that Hamilton rejects the Scripture as revelation. He says,
“Were this the case [identifying the words of the Bible with the
word of God] then the Bible, rather than being that inspired
record ... would be the written law of God” (p. 76).
Now there is asense in which the Bible is an inspired record. It
inerrantly records God’s revelation to Abraham and the wars of
David King of Israel. But in addition to being a record of divine
revelations, it is itself the complete revelation. As the opening
section of the Westminster Confession (determinative of the
evangelical position) says, “it pleased the Lord . . . tocommit the
same [earlier revelations] wholly unto writing . . . those former
ways of God’s revealing his will unto his people being now ceased.”
Thus in contrast with Hamilton’s denial, the Bible is indeed the
written law of God.
In spite of the fact that Hamilton wants to escape myth through
poetry to parable, he continues to say, “The language of Scrip-
ture .. . would have been incomprehensible otherwise . . .”(p. 89)
i.e., unless mythic patterns had been used. Ananias in Damascus
would not have understood the directions to Straight St., had it not
been mythological in form. “Sumerian, Babylonian, Phoenician,
and Egyptian myths[were] taken up into the biblical accounts of
creation” and “Gnostic myths [are] present in the N.T. descrip-
tions of Christ.2. .. The biblical language employs the imagery
of myth, while transforming its content.4 Creation myths in which
the gods wrested apart earth and heaven out of the body of the
monster Chaos account for some of the phrasing of the biblical
account of creation” (p. 89).
Clearly, however much Hamilton may want to go beyond myth,
he does not seem to get very far away, for on the next page he says,
“Lacking the mythic pattern [of Gosticism] that originally
produced the necessary terminology, we should not be able to
speak of Christ’s death and resurrection” (p. 90).

’For a definitive refutation, see The Origin of Paul’s Religion, J. Gresham

Machen.
4Does it? How? With what result?
KENNETH HAMILTON ei

Is this not complete nonsense? Am I dependent on Gnostic or

other myths when I speak of Roman soldiers laying Jesus on a
cross and pounding nails into his hands and feet? Certainly I
understood this in childhood long before I ever heard of Gnosti-
cism. Nor am [| at all sure that Matthew knew anything about
Gnosticism. If anyone now replies that Matthew and I did not
need to have known Gnosticism because we use language already
formed, let him explain to us how mythology formed the words:
nails, soldiers, cross, spear, and death. Similarly, what mythology
is needed for Peter to see that the tomb was empty and later to see
Jesus in Galilee and talk with him? Is it not therefore complete
nonsense to say that we could not talk about Christ’s death unless
mythology had given us these words?
One hardly escapes the impression that the author does not
treat his opponents fairly. He says, “Yet because revelation is
given in human words, it cannot be more precise than language
allows. {How true! A perfect tautology. But is God, who produced
language, unable to use it with perfect precision? ]The belief that
the Bible consists of statements of literal truth [ital. his], therefore,
is ill-conceived. [The therefore is a logical fallacy.] The notion of
literal truth is quite correct if we oppose literal to the mythi-
cal....In this sense we must say that God literally created the
world... . It is quite another matter, though, if we insist that all
the statements of Scripture are literally true...” (p. 91). This
sort of argument is hardly fair to the Reformation view because
no one from the time of Moses to the present ever said that all
statements are strictly literal. Did Luther, Quenstedt, Gaussen,
or Warfield ever say so? Of course there are figures of speech,
metaphors, anthropomorphisms, and the like. But these would be
meaningless if there were no literal statements to give them
meaning. For example, II Chronicles 16:9, “The eyes of the Lord
run to and fro throughout the whole earth,” is ludicrously ridicu-
lous if taken literally: little eyeballs rolling over the dusty ground.
But unless the statement, God is omniscient, is literal, the figure
has nothing to signify. Surely Hamilton did not publish his book
to remind us that the Bible contains some figures of speech. And
yet his argument here depends on the alleged fact that someone
said “all the statements of Scripture are literally true” (p. 91).
Consider the footnote on page 91: “ ‘Literal’ is not synonymous
128 KENNETH HAMILTON

with ‘historical.’ Inspiration does not imply that what is inspired

must be understood literally, and even less that everything must
be viewed as having actually happened. .. . To put it bluntly, to
accept everything reported in the Bible as having actually hap-
pened, one must tamper with the text.” These words which
Hamilton with approval quotes from H. M. Kuitert are unclear.
The language is typical of liberals who want to appear conserva-
tive to orthodox people, while they undermine the truth of the
Scripture. When Kuitert says “everything reported,” does he
refer to metaphors, to statements made by Satan, or does “every-
thing reported” refer to everything reported as having actually
occurred? The first two possibilities are puerile. The third is a
repudiation of evangelical religion. It is hard to avoid the conclu-
sion that the latter is the meaning intended. For example, II Peter
claims that it was written by Peter. About such aclaim Hamilton
writes, “For along time now, every author has been considered to
have a proprietary right over his works. But the biblical books
came out of a milieu in which such a concept was unknown, and
where there was no issue of truth or falsehood involved in using a
revered name in connection with writings by other hands” (p. 92).
This statement is not true even of pagan scholarship, for the
Alexandrian philosophers carefully distinguished between thirty-
six genuine Platonic dialogues and ten spurious. See also E. M. B.
Green, Second Peter Reconsidered (Tyndale Press, 1960), where
he writes to the effect that forgeries were not cordially received
as the critics maintain, but that the subapostolics distinguished
themselves and even Apollos from the apostles, and deposed the
author of Paul and Thekla for his imposture. Another instance was
Serapion, who banned the Gospel of Peter from his church because
by careful investigation he had discovered it was a forgery.
After his remarks on the authorship of spurious writings,
Hamilton comes quickly to his solution to the problem of how
language with its mythical inheritance can express divine truth.
It is done by parable. The book of Jonah, he says, does not report
actual occurrences. Its literary form shows that it is a parable.
[There never was a Jonah. I guess there was no Ninevah, either.]
Everyone acknowledges that Christ taught in parables.* Not

*A common criterion for distinguishing a parable of Christ from something he

KENNETH HAMILTON 129

everything in the Bible, Hamilton acknowledges, is a parable; the

apocalyptic visions are not. But “if we are to look for a ‘key’ mode
of language-usage in Scripture, then parable fits this position
much more suitably than myth does” (p. 100).
Let us immediately agree. There are also other sentences in the
book, which, if detached from their context, can be understood in
an orthodox sense. So, it is true that parable is more suitable than
mythology. But is parable more suitable than, and a substitute
for, literal language? Hamilton has made the wrong comparison.
He has here avoided mentioning the weak link in his argument;
for if there is no literal truth of which the parable is an illustration,
it has no referent and becomes pointless.
It would indeed seem that Hamilton has made parables point-
less and meaningless. He says, “A parable... assumes that the
divine reality its human words open to us, though literally beyond
our comprehension, can actually be revealed to us by means of
human words. Thus many of the parables of Jesus begin, ‘The
kingdom of God is like. .. .. Certainly the comparison is no more
than a comparison. The kingdom of heaven cannot be brought
down to earth for our inspection; it remains alwaysa mystery. Yet
Jesus could say... ‘It is given unto you to know the myster-
WSs c Ap. OG):
This quotation is peculiar. It begins by saying that the sense of
the parable, that is, the divine reality it reveals, is literally beyond
our comprehension, but ends with Christ’s assertion that the
disciples should understand it. In the middle is the word mystery:
the kingdom remains always a mystery. But mysteries are not
necessarily impossible or even difficult to understand. In the
N.T. mystery does not refer to something we call mysterious in
English. For example, I Corinthians 15:51 states a mystery: it
may be hard for some people to believe, but there is no difficulty in
understanding it.
Then, too, it is false to say that “the kingdom of heaven cannot

reports as having happened is the absence in the first and the presence in the
second of names: a man that was a householder went out early to hire laborers, ora
certain king made a marriage feast for his son, versus the blood of Abel. . . of
Zachariah, son of Barachiah, whom ye slew, or other references to O.T. events.
 130 KENNETH HAMILTON

be brought down to earth for our inspection.” Christ did just that.
Also the kingdom remains with us, and we inspect it daily.
But once more, if “the comparison is no more than a compari-
son,” or, better, if it is as much as a comparison, the particular
truth illustrated by the comparison must be understandable, for
otherwise the parable’s language would not reveal the truth to
us.
In conclusion, first, Hamilton’s theory of language is destruc-
tive of Christian truth. Surely language, as God’s gift to Adam,
has as its purpose, not only communication among men, but
communication between man and God. God spoke words to Adam
and Adam spoke words to God. Since this is the divine intention,
- words or language is adequate. To be sure, on occasion, even on
frequent occasions, sinful man cannot find the right words to
express his thought; but this is a defect of man, not an inadequacy
of language. The Bible does not countenance a theory that origi-
nates language in pagan mythology with the result that divine
truth is unintelligible.
Similarly, second, on Hamilton’s theory God remains unknow-
able. The chief difficulty with myths is not that they are literally
false, but rather that their alleged nonliteral “truth” is meaning-
less. Hamilton fled from myth to poetry to parable in order to
arrive at some sort of revelation, but he never succeeded in
showing how parables convey truth or what truths parables
convey. Their “message” remains unintelligible.
Third, Hamilton has rejected the doctrine of verbal and plenary
inspiration and places himself outside the bounds of historical
evangelicalism.
 8
A Christian Construction

To prepare for a positive formulation of a Christian theory of

language, the first thing is to clear the ground of empiricism.
Many Chris<‘an evidentialists, unwilling to accept liberal or neo-
orthodox positions, are nonetheless unwilling also to hack away
and dig out the roots of non-Christian branches of learning. When
a nonempirical apologetic is presented to them, they almost
always reply with the boldest and most naive petitio principit:
‘Don’t you have to read the Bible?’
A serious apologist cannot ask this question until after he has
defined sensation and explained its relation to perception. A polo-
getics or Christian philosophy has the task of formulating a
complete and consistent theory from beginning to—if not end, at
least as far as one can go. But it must start at the beginning. When
someone asks, ‘Don’t you read your Bible?’ he is assuming that a
Bible is certain sensations of black and white without combina-
tion, arrangement, or intellectual interpretation. Now, this is
clearly not the case. The perception of a Bible is somehow ordered
and interpreted. The apologist must explain how. An empirical
appeal, like the sight of a Bible, cannot be the beginning of an
epistemological theorv. If the apologist cannot show how percep-
tion of a Bible develops from sensation, he has no basis for his
empiricism. He has no defense against a spiritual rationalism.
The former, so we apprehend, lands him either into behaviorism
or chaos—which are much the same thing. The latter provides for
an intelligible message from God.
Even some nonempiricists try to connect sensation with per-
ception. Brand Blanshard, an Hegelian, not a Christian, in his

131
132 A CHRISTIAN CONSTRUCTION

brilliant volumes on The Nature of Thought makes an excellent

attempt. In chapter one, paragraph three, he states, “Perception
is that experience in which, on the warrant of something given in
sensation at the time, we unreflectingly take some object to be
before us.” Note the words given and unreflectingly. Here is das
Gegebenes by which Hegel pointed out the fatal flaw in Kantianism.
Now, Blanshard works out his arguments with great care. But
after one has reflected on the first seventy-five pages, it may
dawn on the reader that, contrary to Blanshard’s intentions,
there is no such thing as sensation. He himself acknowledges that
no adult has, or remembers as a baby having had, a sensation.
Thus there is no empirical evidence of sensation. Second, if there
were, and if perception were an inference from sensation, it
would be necessary to show which inferences are valid and which
are invalid. Blanshard very carefully gives many examples of
possible inferences; but he nowhere shows how to distinguish a
true perception from a fallacious inference. If Blanshard has not
done this, no wonder Christian evidentialists have not. Usually
they have never even thought of the problem. To repeat: an apolo-
getic system must begin at the beginning, not half-way along the
road; and from the beginning every step must be validated. The
challenge to the Christian evidentialist is: Can you from the
sensation of blue—if blue is really ein Gegebenes and not already a
universal—validly infer anything whatever? Apologetics must
be systematic. How then can the doctrine of the Trinity be based
on the visual sensation of blue or the tactual sensation of pain?
In the earlier analyses of logical positivism and similar views,
some other arguments against empiricism have been indicated.
For full discussions the reader must consult the 2500 year history
of philosophy. Yet a few samples, if seriously considered, suffice
for the present purpose.
First, sensation or perception is untrustworthy. What one
person sees as red and green, another sees as two shades of gray,
and maybe not even as two shades. One focus makes a small near
object, if such is the case, appear asa far large object. If a piece of
canvas is painted half red and half green, astroke of gray through
the two backgrounds produces two different color sensations. To
generalize, every sensation has a background; nothing can be
seen alone. This background is one of the factors that give the
A CHRISTIAN CONSTRUCTION 133

object its observed qualities. Hence the Bible in your hand is not
itself black; and if we do not know what color it is, how can we
know it is a book at all? This question must not be ignored—if one
wishes to be an empirical apologist. Remember the Texan rancher
who was sure he was seeing a mirage and drove his jeep into a lake.
The same difficulty recurs in every sense. Anexample for taste
would be the difference between drinking grapefruit juice and
then eating ice cream, and eating the ice cream first and then
drinking the grapefruit juice.
Though Thomas Aquinas did not seem troubled by these points,
even he acknowledged that the sense organ had to be healthy in
order to perceive the real color or taste of the object. But who
knows whether his sense organs are in perfect condition or not?
Even when medical tests discover no disease in the sense organs
of two persons, must we say that one person is sick because he
does not like his taste of onions and the other is well because he
gets a pleasant taste from them?
Differences in taste, optical illusions, tricks in perspective,
color blindness, and then the too definite and too manifold dis-
tinctions in color that other people call hallucinations, fill ele-
mentary textbooks on psychology. The apologist who wants an
empirical theory of language or an evidentialist defense of the
faith ought to examine the foundation on which he wishes to rear
a superstructure.
It is interesting to note what Calvin has to say on these subjects.
To be sure, he was not a philosopher, but a good theologian must
have some opinions on epistemology. The following quotations
will show that he was not exactly an empiricist.
“The eye, accustomed to seeing nothing but black, judges that
to be very white which is but whitish or perhaps brown. . . If at
noonday we look either on the ground, or at any surrounding
objects, we conclude our vision to be very strong and piercing;
when we raise our eyes to the sun, they are at once dazzled and
confounded ...and we are constrained to confess that our
sight ... is dimness itself” (Institutes, I, i, 2).
“The powers of the soul are far from being limited to functions
subservient to the body. For what concern has the body in measur-
ing the heavens, counting the number of the stars, computing
134 A CHRISTIAN CONSTRUCTION

their several magnitudes, and acquiring a knowledge of their

respective distances. . . . Inthese profound researches relating to
the celestial orbs, there is no corporeal cooperation, but that the
soul has its functions distinct from the body” (zbid., I, v, 5).
There is more. A paragraph or two ago, the theory that percep-
tion is an inference from sensation was found wanting. But ifone
tries to escape this inference theory, he faces a harder difficulty.
It is this. At any one time a person has impressions of red, smooth,
sweet, and dozens of others. To perceive a thing, these ‘sensations’
must be combined. Note that no one even sees a dog or a tree. A
dog is not just black; he is also soft, fuzzy, and perhaps has an
odor. But before one perceives a dog, he must choose black, fuzzy,
and odor, combine them, and only then has he the perception of
his pet. Yet there is nothing in the single qualities that forces him
to select these particular ones and discard the dozens of others he
also has at the same time. Why does he not select the fuzzy, the
sound of B flat, and the taste of Bacardi rum, all of which he
senses at the same moment, and combine them into the perceived
object? Is there anything in a person’s fifty or more sensations
that compels the selection of these few rather than another few?
Usually people say that they combine the sensations emanating
from the same place. Well, aside from the difficulty of locating
the particular spot from which an odor, or sound, emanates, this
answer presupposes a knowledge of space in general. Where,
then, did the knowledge of space come from? Has anyone seen,
smelled, or touched it? Kant tried to defend a knowledge of space
against Hume; but he could not remain an empiricist to do so. He
had to have apriori forms of the mind.
The next difficulty, and with this one we may need no more to
rid ourselves of empiricism, is the formation of concepts. The
theory is that perceptions produce images which remain after the
perception ceases. By a process of abstraction, concepts are
formed or extracted out of these images.! There are two impossi-

'D. C. Dennett in Content and Consciousness (New York: Humanities Press,

1969, pp. 182-146) seems to be arguing that no one ever has any mental imagery.
Much of what he says is hard to understand. One trouble is that he makes
assertions that seem prima facie false. For example, “We can do without the
dimensionlessness of mental images (that strange quality that prevents us from
A CHRISTIAN CONSTRUCTION 1385

bilities here. First, the theory assumes that all people have such
images. Did not Russell say that only amad man could deny it? So
did Hume. Once again Brand Blanchard not only shows the
futility of images for people who have them, but also brings
before us a group of scientists and literary men, all well educated,
who have no images whatever. Further, the present writer’s
investigations over a long span of years completely confirm the
point. But second, the process by which concepts are allegedly
abstracted from images is unintelligible. Aristotle simply gives
an analogy. It is like an army in rout: one soldier makes a stand,
then a second, and so on, until the army is in order again. This
analogy is worse than most. It is unintelligibility raised to an
unimaginable power.
Christianity, however, must have what most people call
‘abstract’ concepts. Empiricism with its nominalism cannot
produce concepts, such as justification, federal headship, or
Trinity. Nor can it produce the concept of the general conic, of
vertebrate animal, or of tennis. To speak more precisely: there
are nosuch things as abstract concepts. Abstraction is impossible.
This leads to another point. When a Christian uses the word
justification, Trinity, or theology, he is using a name to designate a
series of propositions. A student does not know botany: he knows
that asparagus and the star of Bethlehem are members of the
liliaceae. To know theology is to know that ‘Adam was the federal
head of the race,’ and that ‘the elect sinner is justified by means of
faith alone.’ Propositions, not concepts, are the objects of knowl-
edge because only propositions can be true.
Theological propositions are usually universal propositions,
and for that reason cannot be empirical. Empiricism is ruled out, not

putting any kind of ruler, physical or mental, along the boundaries of mental
images) and their penchant for inhabiting a special space of their own, distinct
from physical space” (p. 141). Now, although I have no memory images myself, I
accept the statements of Aristotle, Hume, and Russell that they do. And not only
these philosophers; but after questioning classes of college students for fifty years,
I am convinced that the great majority of people have visual imagery. A lesser
percentage has auditory and tactual images. Hence, to reply to Dennett I canonly
ask this majority if they have ever dreamed of measuring something with aruler.
How Dennett can say that this is impossible is something I do not understand. His
underlying behaviorism, however, is no subject for a short footnote.
136 A CHRISTIAN CONSTRUCTION

because these propositions are matters of revelation, undiscover-

able by an unaided human mind, but because they are universal.
“All who are justified are justified by faith alone” is a universal
proposition. But induction never arrives at universals. And
induction is all that empiricism has. By induction a young orni-
thologist may observe a thousand black crows—not to repeat all
the difficulties of seeing even one black crow—and on the basis of
these thousand observations he is likely to assert ‘All crows are
black.’ Then the thousand and first crow is an albino. Induction
never arrives at a universal. If so used, it is always a logical
fallacy.
Empiricism, therefore. is in asad state, so that not much can be
said in favor of a language theory or a theology based on it.
Finally, now, the constructive theory can begin. If the secularist
doubts that there is or can be a Christian theory, the Christian
may doubt that the Bible discusses language at all. What verse or
chapter defends metaphysics? What book gives us any verification
principle? The Psalms indeed are poetry, but do they tell us that
poetry is an advance over myth on the way to parables? Do they
even teach otherwise? Though many believe that the Bible is
silent on these philosophical matters, the following is an attempt
to show that the Bible answers all these questions, and with
reasonable clarity.

The first part of the answer, the first element in the formulation
of aChristian theory of language, and therefore the first criterion
for judging the adequacy of biblical revelation, is the doctrine of
the image of God in man. Or, rather, the very first part is the
biblical doctrine of God. Is God the ‘Totally Other’? Do God and
the medium of conceptuality “schliessen einander aus,” com-
pletely exclude each other? Or is God an object of thought and
knowledge as much as or even more than the square root of minus
one?
In thinking about God Calvinists almost immediately repeat
the Shorter Catechism and say, “God is a spirit, infinite, eternal,
and unchangeable.” Perhaps we do not pause to clarify our ideas
of spirit, but hurry on to the attributes of “wisdom, holiness,
justice, goodness, and truth.” But pause: Spirit, Wisdom, Truth.
Psalm 31:5 addresses God as “O Lord God of truth.” John 17:3
A CHRISTIAN CONSTRUCTION 137

says, “This is life eternal, that they might know thee, the only true
God... .” I John 5:6 says, “the Spirit is truth.” Such verses as
these indicate that God is a rational, thinking being, whose
thought exhibits the structure of Aristotelian logic.

If anyone objects to Aristotelian logic in this connection—and

presumably he does not want to replace it with the Boolean-
Russell symbolic logic—let him ask and answer whether it is true
for God that if all dogs have teeth, some dogs, spaniels, have
teeth? Do those who contrast this ‘merely human logic’ with a
divine logic mean that for God all dogs may have teeth while
spaniels do not? Similarly, with ‘merely human’ arithmetic: two
plus two is four for man, but is it eleven for God?

The verses quoted in the next to the last paragraph are only a
small fraction of all those which teach the rationality of God. The
mass of material that asserts or implies God’s omniscience,
material to which Charnock devoted a good two hundred pages,
would, if repeated here, impress those who have unfortunately
never read Charnock, but would doubtless seem tedious to others.
Its shortest summary is I Samuel 2:3, “The Lord is a God of
knowledge.”

Special mention, however, should be made of God’s seeing the

end from the beginning, his doing all things well, his moral
governance of angels and mankind, his choice of means (some-
times surprising) to accomplish his plans, and, of course, the
plans themselves. From among the multitudinous instances let
us consider only II Chronicles 14-20; and from these chapters
only two texts: 16:9, “The eyes of the Lord run to and fro through-
out the whole earth to show himself strong in behalf of those
whose heart is perfect toward him”; and 18:19-221, “The Lord
said, Whoshallentice Ahab. .. .Thenthere came out aspirit and
stood before the Lord and said, I will entice him... . And the
Lord said, ... Thou shalt also prevail: go out and do even so.”
Here the Lord is said to have had and to have executed a rational
plan to destroy Ahab. If, now, God works all things against his
enemies, for our good, and for his glory, we may confidently insist
that God is a rational being, the architecture of whose mind is
logic. Irrationality contradicts the biblical teaching from begin-
ning to end. The God of Abraham, Isaac, and Jacob is not insane.
138 A CHRISTIAN CONSTRUCTION

With this understanding of God’s mind, the next step is the

creation of man in God’s image. The nonrational animals were
not created in his image; but God breathed his spirit into the
earthly form and Adam became a type of soul superior to the
animals.
To be precise, one should not speak of the image of God in man.
Man is not something in which somewhere God’s image can be
found along with other things. Man is the image. This, of course,
does not refer to man’s body. The body is an instrument or tool
man uses. He himself is God’s breath, the spirit God breathed into
the clay, the mind, the thinking ego. Therefore, man is rational in
the likeness of God’s rationality. His mind is structured as Aris-
totelian logic described it. That is why we believe that spaniels
have teeth.
In addition to the well known verses in chapter one, Genesis 5:1
and 9:6 both repeat the idea. I Corinthians 11:7saysthat“man...
is the image and glory of God.” See also Colossians 3:10 and James
3:9. Other verses, not so explicit, nonetheless add to our informa-
tion. Compare Hebrews 1:3, Hebrews 2:6-8, and Psalm 8. But the
conclusive consideration is that throughout the Bible as a whole
the rational God gives man an intelligible message.”
This intelligible message not only includes accounts of histori-
cal events and God’s explanation of them, but also a number of
commandments and precepts. Though the gift of rationality was
an inestimable blessing, it also carried danger with it. Because
animals are nonrational, they cannot sin. Man could and Adam
did. The very possibility of sin depends on a law that God imposes
and that man can understand.
This point brings us to the central issue of language. Language
did not develop from, nor was its purpose restricted to, the
physical needs of earthly life. God gave Adam a mind to under-
stand the divine law, and he gave him language to enable him to
speak to God. From the beginning language was intended for
worship. In the Te Deum, by means of language, and in spite of
the fact that it is sung to music, we pay “metaphysical compli-

2Cf. Gordon H. Clark, “The Image of God in Man,” J.E.T.S. XII, iv, 1969.
A CHRISTIAN CONSTRUCTION 139

ments” to God. The debate about the adequacy of language to

express the truth of God is a false issue. Words are mere symbols
or signs.
Urban distinguished between signs and symbols. If this dis-
tinction be accepted, words are signs. Even if his onomatopoeic
words are symbols, it does him little good, for there is no inherent
quality in the sound of dog, chien, or Hund to make it mean a
certain type of animal. Any sign would be adequate. The real
issue is: Does a man have the idea to symbolize? If he can think of
God, then he can use the sound God, Deus, Theos, or Hlohim. The
word makes no difference; and the sign is ipso facto literal and
adequate.
Those who resort to myth, parables, or imagery—pictorial
symbols used for inconceivable objects—and who declare that
language is inherently inadequate and the mind inherently
incompetent to speak about God, deny that God is able to attach
signs to thoughts and to create a mind that can understand the
thoughts. The god of Gilkey and Hamilton is not omnipotent; he is
not God Almighty; he is powerless to give man any understanding
of himself. Their god is unable to speak the truth to Abraham, nor
can Abraham address himself intelligibly to him.
The Christian view, on the contrary, is that God created Adam
as arational mind. Thestructure of Adam’s mind was the same as
God’s. God thinks that asserting the consequent is a fallacy; and
Adam’s mind was formed on the principles of identity and con-
tradiction. This Christian view of God, man, and language does
not fit into any empirical philosophy. It is rather a type of a priori
rationalism. Man’s mind is not initially a blank. It is structured.
In fact, an unstructured blank is no mind at all. Nor could any
such sheet of white paper extract any universal law of logic from
finite experience. No universal and necessary proposition can be
deduced from sensory observation. Universality and necessity
can only be apriori.
This is not to say that all truth can be deduced from logic alone.
The seventeenth century rationalists gave themselves an impos-
sible task. Even if the ontological argument be valid, it is impos-
sible to deduce Cur Deus Homo, the Trinity, or the final
resurrection. The axioms to which the apriori forms of logic
must be applied are the propositions God revealed to Adam and
140 A CHRISTIAN CONSTRUCTION

the later prophets. Logic is stressed here, however, because

empiricism not only makes revelation impossible: it makes
thought and truth impossible. Liberal religion, therefore, must
be mythological because, even worse, it must be irrational. And if
irrational, of course there can be no meaningful revelation. Nor,
for that matter, can there be any geometry. Geometry needs the
universal and irreplaceable laws of logic, and these empiricism
cannot provide.
Logic is irreplaceable. It is not an arbitrary tautology, a useful
framework among others. Various systems of cataloging books in
libraries are possible, and several are equally convenient. They
are all arbitrary. History can be designated by 800 as easily as by
500. But there is no substitute for the law of contradiction. If dog
is the equivalent of not-dog, and if 2 = 3 = 4, not only do zoology
and mathematics disappear, Victor Hugo and Johann Wolfgang
Goethe also disappear. These two men are particularly appro-
priate examples, for they are both, especially Goethe, romanti-
cists. Even so, without logic, Goethe could not have attacked the
logie of John’s Gospel (I, 1224-1237).

Geschrieben steht: “Im Anfang war das Wort!”

Hier stock ich schon! Wer hilft mir weiter fort?

Mir hilft der Geist! Auf einmal seh’ ich Rath

Und schreib’ getrost: “im Anfang war die That!”

But Goethe can express his rejection of the divine Logos of John
1:1, and express his acceptance of romantic experience, only by
using the Logic he despises.
To repeat, even if it seems wearisome: Logic is fixed, universal,
necessary, and irreplaceable. As such its laws cannot be deduced
from nor abstracted from experience. If dog, cat, typewriter, as
well as Wort, Geist, and That, all mean the same thing—as they
must apart from the law of contradiction, empiricism can express
nothing: Goethe and Racine, Hegel, and Kierkegaard, are the
same person.
Christianity, on the other hand, requires and justifies universal
propositions as well as distinguishing Ahab from Jehoshaphat.
The doctrine of ‘he Atonement, for example, that Christ’s death
A CHRISTIAN CONSTRUCTION 141

was a satisfaction for sin, is a fixed truth and cannot be replaced

by its contradictory. To bring this point quite up to date, the
teachings of Scripture are not “culturally conditioned” so as to
have been “true” in antiquity, but antiquated today. Similarly,
the Ten Commandments—in this decadent age, an acceptance of
premarital sex and homosexual practice keeps company with a
hue and cry against the ‘dictation theory’ of inspiration. But what
is so bad about dictation? Luke, of course, acknowledges that he
made investigations; and it is not supposed that God dictated the
wording of the birth narratives. But how else could Moses have
written the Ten Commandments? There were not so much dictated
as written out on stone by God’s omnipotence. Note, too, that the
first chapter of Genesis was not the result of historical or geologi-
cal investigation. The genealogies inI Chronicles may have been
copied from earlier documents; but in the case of Isaiah God must
have mentioned Cyrus by name. Impotent gods may not be able to
speak; but Jehovah spoke. Christianity is based on revelation, not
experience.
Since God is both rational and omnipotent he faced no problem
in adequately expressing his truth in words. Because man is also
rational, he faces no inherent problem in understanding God’s
words. Since the fall, indeed, he is often likely to misunderstand.
Heis, however, more likely to understand quite well, but refuse to
believe. Both misunderstanding and refusal to believe are to be
classified under the pedantic category of the noetic effects of sin.
The first may sometimes be attributed to the incompetence of the
Christian preacher. But neither is the result of any inherent
inadequacy of language.
As for the language of proclamation, the central truths of the
gospel can be expressed simply. It is also legitimate, even on
many occasions, to use a more recondite literary style. God’s
prophets, if they do not use mythology, nonetheless use metaphors,
poetry, and parable. These have their literary value. But their
meaning can always be expressed in straightforward prose.*
Perhaps the problem of communication has here been some-
what neglected in the effort to defend logical thinking. But if

3Cf. Religion, Reason and Revelation, pp. 144ff.

142 A CHRISTIAN CONSTRUCTION

symbols are always adequate simply because the thinker chooses

an otherwise meaningless sound or mark to designate his idea,
there remains the difficulty of communicating the idea by means
of the symbol to another mind. This is a point dear to the heart of
Christian empiricists. ‘Don’t you read your Bible?’ they ask. ‘Don’t
you see the words on the page?’ Now, these questions deserve an
answer, and it shall be given. But note first that the empiricist
has a harder time explaining communication than the rationalist
or intellectualist has. How can sensationism produce a sound that
conveys a meaning from one mind to another? Since my sensation
is never yours, how can you ever know what the sensation is to
which I attach a sound or ink mark? The empirical apologists
usually evade this problem.
St. Augustine, with his Platonic background, did not evade.
His discussion constitutes the second half of his tractate De Mag-
istro. The good bishop showed, conclusively I should say, that the
ostensive definitions of logical positivism’s protocol sentences are
failures. His solution was, briefly, not that two minds had the
same sensations, but that two minds have the same ideas. The
ideas are common because Christ is the Logos that lighteth every
man that comes into the world. “In him we live and move and
have our being.” Malebranche, perhaps not to be followed in
every detail, for no one is, used the figurative phrase, ‘We see all
things in God.’ Perhaps a modern example will prove useful.
Before World War II, Japan, Great Britain, and the United
States discussed naval limitation. The eryptographers in Wash-
ington got possession of Japanese documents written in code.
Even though the cryptographers knew no Japanese, they were
able to decipher the coded information. The reason is that logic is
the form of every human mind. The Japanese may not use the
Arabic numbers 5-5-3; but whatever characters they use, they
cannot use the same character for every number. Similarly,
Champollion deciphered the Egyptian hieroglyphics. And be-
tween the ages of one and two a baby’s logical mind can decipher
whatever symbolism his mother uses. It is only college students
who have trouble with German, French, or Greek, and dialectical
theologians with English.
This much should be sufficient for the subject of communica-
tion. The general Christian public, however, who do not hold
A CHRISTIAN CONSTRUCTION 143

doctorates in philosophy, are more interested in an exegetical

problem connected with the question previously mentioned: ‘Don’t
you ready your Bible?’ Dr. Robert L. Reymond of Covenant
Seminary in St. Louis, one of two critics who have summarized
the position here maintained with commendable accuracy, puts
the problem in its clearest terms. He writes,
There are scores of biblical passages which teach by inference, if
not directly, that sensory experience plays a role in knowledge
acquisition, (e.g., Matt. 12:3, 19:4, 21:16, 22:32: Mark 12:10; Rom.
10:14). It seems to me, before he will convince many Christians of
his position, that Clark must explain satisfactorily (in another way
than is virtually universally taken) literally hundreds of passages
of Scripture which employ the words'‘see,’ ‘hear,’ ‘read, ‘listen,’ etc.
At this time I am not convinced that he is in accord with Scripture
when he denies to the senses a role in knowledge acquisition and
would hope that he would take the Greek skeptics less seriously
and the implications in many ofthe ‘subsidiary axioms of Scripture
more seriously than he does.
Two pages earlier he cites I John 1:1-3, which is perhaps more
pointed than the others, for it says, “That which... we have
heard... seen with our eyes, ...our hands have handled...
that which we have seen and heard, declare we unto you.” Do not
these words guarantee that Christianity is a form of empiricism,
a system based on experience?
Now, I am willing to exegete such verses, and I shall do so,
briefly here and more at length in a commentary on I John that
should appear shortly. But first there are one or two minor
phrases in Reymond’s paragraph that call for notice. His words
“denies to the senses a role in knowledge acquisition” are vague,
for they do not specify what role. Animals have more acute sensa-
tions than human beings; but they know no mathematics, con-
struct nosyllogisms, nor do they write narratives. Sensation does
not help them in these matters. Sleeping and eating play a role in
knowledge acquisition in this life, for without them we would not
remain in this life. But their role contributes nothing to the
content of knowledge. Nutrition plays a role, but it is not true that

4Robert L. Reymond, The Justification of Knowledge (Nutley, N.J.: Presbyterian

and Reformed Publishing Co., 1976), p. 114.
144 A CHRISTIAN CONSTRUCTION

“Der Mensch ist was er isst.” Philosophers who insist on giving a

role to sensation in the acquisition of knowledge should first
define sensation, then show how sensation can become perception,
and presumably how memory images can produce universal
concepts by abstraction. If this is not their scheme, and it might
not be, then they should describe in detail what their scheme is. It
is not enough to speak vaguely about some role or other. Plato
gave the senses the role of stimulating reminiscence. Presumably
this role would not satisfy Dr. Reymond. St. Augustine, though he
altered his views as he grew older, gave a different role to sensa-
tion; without too much distortion one might call it a stimulus to
intellectual intuition. Would that satisfy Dr. Reymond? It is hard
to say because Dr. Reymond himself does not give any role to
sensation. No doubt he believes that there issome such role, but I
must have missed the page on which he tells what that role is.
Now, it is not necessary for a critic to explain his own view in
order to reject the view he is criticizing. But if one writes on The
Justification of Knowledge, the readers expect a specific explana-
tion. This ties in with the second defect in the paragraph quoted.
He thinks that I take the Greek skeptics too seriously. Of course, it
is not the Greek skeptics alone that I take seriously. There are also
Montaigne, Descartes, Bayle, Hume, and the contemporary
experiments in psychology. It would be my desire that Dr.
Reymond, with his considerable ability, might take all skepticism
more seriously. Responsibility to the task of apologetics demands
it. Unfortunately several conservative apologists, with whose
theological views I am in substantial agreement, seem to me to
have evaded this basic problem. It has been stated clearly in this
monograph, and I cannot believe that it should not be taken
seriously. Just one more minor point: Dr. Reymond’s disagree-
ment with my reply to Dr. Nash(pp. 112-113) omits one essential
fact: the fact that Dr. Nash does not correctly report my view. He
asserts that I hold, “Man cannot know the contents of the Bible
save through the senses.” If I am correct in assuming that
Reymond and Nash both reject the view that a sensation can be no
more than astimulus to recollection or intellectual intuition, then
Nash does not correctly state my view, and hence his deductions
from this statement are inapplicable to me.
However, we must get closer to exegesis. Before examining
I John 1:1-3, it may be well to note that the word sensation
A CHRISTIAN CONSTRUCTION 145

(aisthesis) occurs only once in the New Testament: Philippians

1:9. Neither KJ, RSV, NAS, nor NIV translate it sensation. It
does not mean sensation. Hebrews 5:14 has ta aistheteria (the
faculties of sensation). Some translators have “senses”; but clearly
the word does not mean senses in the sense used in discussions on
sensation. Dr. Reymond’s book does not explain a theory of lan-
guage; and I would be the last to assign to him a view of language
he does not hold. I only surmise that he rejects the theory of
ordinary language, by which meanings are fixed by usage, for he
seems to use the words see, hear, sense without considering how
they are used in ordinary and scriptural language.
What did the Apostle John mean when he spoke of seeing with
the eyes and handling with the hands? Did he mean aisthesis,
proper sensibles, common sensibles, sensation per accidens, or
what?>
In Greek the first word of I John designates the Word of life,
who in verse 4 is identified as Jesus Christ. Since the Epistle and
the Gospel have the same author, it is permissible to connect this
Word of Life with the Word of John 1:1. And no one should object if
we equate this Word with him whom Paul calls “the Power of
God” and “the Wisdom of God.” This second person of the Trinity
is the subject of John’s declaration. Can this eternal Wisdom be
heard with the ears, seen with the eyes, and handled with the
hands? Is the second person of the Trinity an object of sense? The
word hearing comes first; seeing comes second. This discussion
will take them in turn.
As for hearing, one should note that no one can ever hear a piece
of music or aline of poetry. Our opponents, who insist on sensation
as the origin of knowledge, cannot well object to an instance taken
from experience. St. Augustine pointed out that to “hear” music
or poetry, one must at least “perceive” the rhythm. But there is no
rhythm in a single sensation. Even beyond perception it is neces-
sary to have memory before a line of poetry can be recognized as
poetry. A single sound has no rhythm or meter. The first sounds
of aline must be remembered until the last sound occurs; note also

5Cf, Martin, Clark, Clarke, Ruddick, A History of Philosophy (New York: F.S.
Crofts and Co., 1941), pp. 161ff.
146 A CHRISTIAN CONSTRUCTION

that the first sound no longer exists when the last sound sounds.
Therefore no one ever senses music or poetry. This Augustinian
remark should satisfy any empiricst; but, of course, it is not
exegesis.
As the noun aisthesis in Scripture does not mean sensation, so,
too, the verb to hear does not do so, either. Exodus 15:14 says, “The
people shall hear and be afraid.” The meaning is that the enemies
of Israel will understand the danger of being defeated in battle.
In Numbers 9:8 someone might want to insist that God spoke in
audible words; but in any case an understanding of the directions
is not found in the vibrations of the air or eardrums. Deuteronomy
1:43 indicates that Moses spoke audible words. Of course, the
people heard. But the verse says they did not hear. What is meant
is that the Israelites did not obey. II Kings 14:11 says that “Ama-
ziah would not hear.” Job 27:9, “Will not God hear his cry?” Other
references also, such as Psalm 3:4, speak of God’s hearing prayers.
Obviously the verb hear does not designate a sensation, for God
has no eardrums to be affected by air vibrations. No sensation is
possible in this case. The verse in Job means, of course, that God
will not favor the hypocrite by granting his petition. Similarly,
Psalm 4:1, with its two instances of the verb hear, has nothing to
do with sensation. The language is figurative.
Deuteronomy 29:4 allows a transition from hearing to seeing.
The verse refers to “eyes to see and ears to hear”; but does it refer
to the sense of sight? The phrase is similar to that inI John, “seen
with our eyes ...and our hands have handled.” The verse in
Deuteronomy says that God did not give the Israelites eyes to see
and ears to hear. Does this mean that the Israelites had no eye-
balls, retinas, and appendages on the sides of the head? It does not
mean even that the Israelites could not literally perceive: “the
Lord hath not given you a heart to perceive.” The language is
figurative and means, perhaps that they did not understand what
God meant, or, more likely, that they understood but refused to
obey. Hence, the language of I John does not necessarily, nor
plausibly, refer to sensation and empiricism.
Genesis 3:5 is not a reference to eyeballs and retina. Genesis
16:4 does not mean eyesight. Even though Psalm 13:3 refers to
death, the word eyes is not literal. Similarly, Psalm 119:18. This
instance cannot possibly refer to sensation, for what is to be
A CHRISTIAN CONSTRUCTION 147

“seen” is completely invisible. Then, most ridiculous of all, ‘the

eyeballs of the Lord, on little feet, run to and fro throughout the
whole earth’ (II Chronicles 16:9).

A most interesting event occurs in Daniel 5:5, which says, “In

the same hour came forth fingers of a man’s hand, and wrote
[Mene, Mene, Tekel, Upharsin] . . . and the king saw the part of
the hand that wrote.” Was this a sensation or an hallucination?
Would it have been valid for Belshazzar to infer that he saw a
physical hand? The astrologers saw the writing; but was this
‘seeing’ a sensation? Did the writing remain visible on the wall
until the Medes broke in and killed Belshazzar? This last question
cannot be answered from the text; but it should be clear that
Belshazzar’s ‘seeing’ was not what modern common opinion nor
certainly modern philosophic opinion calls sensation.
Next consider a few verses from the New Testament. Acts 2:27,
31 says, “Neither wilt thou suffer thine Holy One to see corrup-
tion... his soul was not left in hell, neither did his flesh see
corruption.” This can hardly be taken as a denial of some color
sensation. Acts 28:26, 27 repeats in Greek the Hebrew phrases of
seeing and not perceiving; closing their eyes lest they should see
with their eyes. How can this refer to sensations of color, for all
visual sensations must be sensations of color and nothing else. In
I Corinthians 1:26 the seeing cannot possibly be a sensation.
Further Scripture references may be added: Job 19:26, “I shall
see God” cannot be understood as sensation, for God is not a
colored body. Jeremiah 1:11, 13, though visions are not the sense
of sight. Genesis 2:19, 11:5, and 31:50 are not about sensations.
Since Moses’ body lay buried on the east side of the Jordan, did
Peter see Moses on the Mount of Transfiguration? And as for
Peter, allow this paraphrase of Matthew 16:13-17: Whom do men
say thatI am?... And Jesus said, .. . Peter, you never arrived
at that conclusion through any empirical investigation: it was
revealed to your mind by my Father. Clearly the verb to see does
not always, perhaps not even usually, refer to sensation.
This must suffice for the hundreds of verses to which Dr.
Reymond alludes. I hesitantly suggest that his exegesis is defec-
tive because of the imposition of an untenable epistemology. But
now I John. As in the Gospel of John 12:40, here, too, there is no
148 A CHRISTIAN CONSTRUCTION

reference to empirical sensations. The object, namely, the Word

of life, the Reason and Wisdom of God, is not a physical object and
cannot be literally seen and handled. It does not have a color, nor
any degree of hardness, wetness, or any quality of touch. Explic-
itly in I John the object is the truth or proposition, “God is light.”
This proposition cannot be seen in any literal sense. Therefore
since words are arbitrary signs, whose meaning is fixed by
ordinary language, the hundreds of scriptural verbs to which
empirical apologists refer, do not support the role of sensation
which presumably—though they are never clear on what this role
is—those apologists desire to give it.
To finish, once and for all, with the question, ‘Don’t you read
your Bible?’ Abraham Kuyper in The Work of the Holy Spirit (1, 4,
p. 57), beginning with a quotation from Guido de Bres, says,
““That which we call Holy Scripture is not paper with black
impressions.’ Those letters are but tokens of recognition; those
words are only clicks of the telegraph key signaling thoughts to
our spirits along the lines of our visual and auditory nerves. And
the thoughts so signaled are not isolated and incoherent, but
parts of a complete system that is directly antagonistic to man’s
thought, yet enters their sphere.” The analogy may still be too
behavioristic, but the main thought is sound.
One or two other points that Reymond makes are also worthy of
mention. I have mentioned that, taking the scriptural truths as
axioms, all knowledge is deducible from them.® In opposition
to this, Reymond and others object that this limits too much the
extent of human knowledge. Reymond argues that if knowledge
is limited to scriptural implications, we know nothing at all. “I
suggest that this would lead to skepticism, if not total ignorance”
(cbid., p. 110). Surely this is remarkable: if we know the Bible, we
know nothing! At the bottom of the page Reymond repeats, “So
where am I left? It would appear with no certain knowledge of
anything!” It would seem to me, contrariwise, that if a theologian
can deduce six hundred pages of theology from Scripture, he
knows quite a lot.

°A technical qualification is that some scriptural truths may be deduced from

others. In such cases the former would not be axioms.
A CHRISTIAN CONSTRUCTION 149

Of course, he does not know everything. On the view here

defended knowledge is indeed limited. But what epistemology
can guarantee omniscience to man? If Reymond will retract this
inference to complete ignorance, I am willing to acknowledge
that some truths he very much wants to know are not obtainable
on my theory.
On the previous page Reymond had suggested that the West-
minster Confession does not restrict knowledge to what can be
deduced from Scripture. What those divines as individuals
believed, I cannot say. There was one seventeenth century writer,
whom unfortunately Iam unable to name, who held it possible to
be infallible on one point and mistaken on others. His example
was the ‘infallible’ knowledge of a ship-captain regarding the
approach to a harbor. This hardly seems correct. But whatever
the Westminster divines themselves thought, and whether some
of them allowed for more extensive knowledge, Calvin limits
knowledge to scriptural truth. In the Festschrift, The Philosophy
of Gordon H. Clark (pp. 92, 410, 486), one quotation from Calvin
is given, and in another of my volumes a second is given. The one
in the Festschrift is, “I call that knowledge, not what is innate in
man, nor what is by diligence acquired, but that which is deliv-
ered to us by the Law and the Prophets.”
Cannot Calvin support his view by the statement of Paul in
Colossians 2:3? “In whom are all the treasures of wisdom and
knowledge hid.” If so, then no one will find knowledge elsewhere.
Note also that the French Confession of 1559 says, “The Word
contained in these [canonical] books . . . is the rule of all truth”
(la regle de toute verite).
The one piece of ignorance that Reymond seems most anxious
to press against my view is knowledge of oneself. Self-knowledge
has indeed been a philosophical ideal ever since Socrates said,
Gnothi seauton. But it is very difficult. Plotinus’ Enneads, the
extreme difficulty of which philosophers all acknowledge, can be
understood as a gigantic attempt to achieve self-knowledge. Even
those who think the ideal is possible of attainment must wonder
whether anyone has succeeded. Now, Dr. Reymond laments that,
on my theory, “Reymond is unknowable to himself and to everyone
else except God” (p. 110). He very correctly and adequately
explains my reasons for saying so. I might add that I would be
150 A CHRISTIAN CONSTRUCTION

delighted to know Reymond myself, for he is a most interesting

and gracious conversationalist. But two factors preclude this
desideratum. First, “Reymond” is not a simple object of knowl-
edge. “Reymond” is a name given to a very lengthy complex of
propositions. On Dr. Reymond’s position it must be possible to
know some of these propositions without knowing others. On his
position, if I dare guess at it, this must be the case. It is only a
guess because he never says who or what he is. So perhaps Dr.
Reymond does not know himself. This is not too surprising. Pen-
dennis did not know himself. Or if this literary reference is not
sufficiently classical, neither did Oedipus Rex. But these are only
irritating ad hominem remarks. Like the Duchess’ little boy, I
only do it to annoy, because I know it teases.
Therefore, second, the Scripture says, “The heart of man is
deceitful above all things, and desperately wicked: who can know
it?” Did Peter know himself when he said, “Although all shall be
offended, yet will not I”? Did Dr. X, who as a young man strenu-
ously championed the inerrancy of Scripture and later asserted
that Paul did not speak the truth in his epistles, know himself?
Did Mr. Y, a good seminary student, know that he would die an
alcoholic? Did tragic Z, a most faithful servant of the Lord for
many years, know that he would be a suicide? Who can know
himself? Maybe God is merciful in not revealing that knowledge
to us.
In addition to the two scriptural references in the previous
paragraph, consider Psalm 136:6. The psalm as a whole extols the
knowledge of God; but in doing so casts doubt on a man’s knowl-
edge of himself. “O Lord, thou hast searched me and known
me. ... Such knowledge is too wonderful for me; it is high, I
cannot attain unto it.” If anyone dislikes this verse, or to put it
more politely, dislikes my use of this verse, he should set down on
paper the knowledge of himself he claims to know, and then
demonstrate conclusively how he obtained that knowledge. Other-
wise, objections to my view are simply begging the question.
The arguments that Reymond and others offer against my
position are often plausible. To most people they sound like plain
common sense. But sense, not to mention common sense, offers
such enormous difficulties that I must be content with my more
limited knowledge.
A CHRISTIAN CONSTRUCTION 151

Furthermore, Reymond himself is not an empiricist and cannot

consistently make use of sensation in constructing his apologetics.
His account of Thomas Aquinas attests to that. Then, with refer-
ence to Francis Schaeffer he writes, “Surely Schaeffer is aware
that what a man observes is dependent on his religious pou sto.
What Schaeffer observes may not be at all what another man
observes” (p. 142). Yes, indeed! But how, then, can Reymond, in
opposition to my detailed arguments, insist on the infallible
giveness of sensation? Note also, “I am not convinced that the
world is so self-evidently the world that Schaeffer sees” (p. 144).
Wonderful! Then, too, Reymond’s appeal to Scripture alone, on
the following page, is much to my liking: “itisScripturealone...
not the observed phenomena.” And when he quotes me, in such a
gracious fashion, against my other good friend Montgomery
(p. 155), I fail to see his consistency. And since I hold him in high
regard, I understand him to have accepted my position in his
concluding paragraph, of which two clauses are: “the authority of
the word of the self-attesting Christ of Scripture is the only
ground sufficiently ultimate to justify human truth claims, and
until his word is placed at the basis of agiven knowledge system,
that system remains unjustified and no truth assertion [none
whatever] within it can be shown to have any meaning at all”
(emphasis added).
If my esteemed colleague—and I do esteem him—wishes to
make Scripture the sole basis of all knowledge, and then add on
something from a different source, his consistency eludes me.
Does he favor a Kantian combination of a priori forms and sensory
content? Does he have twoa priori forms of receptivity and twelve
for spontaneity? This is another way of asking whether he can
construct an integrated system. Similarly, he must provide a
théory of language that not only preserves biblical inspiration,
but also shows how black marks on white paper give us the
doctrine of the Trinity. Until he does so, he has no basis for
rejecting other views.
Now, for avery brief summary or conclusion, we might return
to the four questions in which Urban set forth the task of language
philosophy: (1) How is language a bearer of meaning? (2) How is
communication possible? (3) What is the relation of logic to lan-
152 A CHRISTIAN CONSTRUCTION

guage? And (4), which seems to be essentially the first question

over again, How can language refer to things?
First, language is a bearer of meaning because words are
arbitrary signs the mind uses to tag thoughts. Second, communi-
cation is possible because all minds have at least some thoughts in
common. This is so because God created man a rational spirit, a
mind capable of thinking, worshipping, and talking to God. God
operates through his Logos, the wisdom that enlightens every
man inthe world. Third, language is logical because it expresses
logical thoughts. Not to deny the noetic effects of sin, examples of
which are incorrect additions and various fallacies in reasoning,
man is still a rational or logical creature and hence he cannot
think three is four or that two contradictories can both be true.
Language therefore is built upon the laws of logic. The fourth
question has the same answer as the first.
Such in brief is the Christian theory of language. On anearlier
page a hypothetical reader asked whether the Bible has any
theory of language, and somewhere the answer was given that
there are verses which at first sight may not be recognized as
such, but which all the same are pertinent. Two of them now
provide an appropriate finis:
Sanctify them through thy word: thy word is truth (John
enya’
Anyone who guards my doctrine (Logos), shall not see death,
ever (John 8:51).
 a 7
‘it~ye GS be -
: 2 a - =<
> ses - ailing Dleds we4 ~s

7
" ee om §°
aie = oti Oe Al oe Sa
=) en a iL mae |
: >) feu, apie. _

et ae : ew —
8 —_ \ Deh. a
- : *~"Un & aim 7
i = . . «=7

Ss 7 — ne
a = a — he

ov - ay 7
a
$4.95

GORDON H. CLARK
Gordon Haddon Clark, Professor
of Philosophy at Covenant College,
Lookout Mountain, Tennessee,
received his Ph.D. from the Univer-
sity of Pennsylvania in 1929. He has
taught at the University of Penn-
sylvania, Wheaton College, and for
28 years was Chairman of the
Department of Philosophy at Butler
University in Indianapolis.
Dr. Clark has written widely in
theology and philosophy, producing
the following titles: Readings in
Ethics (with T.V. Smith), 1931,
1935; Selections from Helienistic
Philosophy, 1940; A History of
Philosophy (with Martin, Clarke, and Ruddick), 1941; A Christian
Philosophy of Education, 1946; A Christian View of Men and Things,
1952; Thales to Dewey, 1957; Dewey, 1960; Religion, Reason and Revela-
tion, 1961; William James, 1963; Karl Barth’s Theological Method, 1963;
The Philosophy of Science and Belief in God, 1964; What Do
Presbyterians Believe? 1965; Peter Speaks Today: A Devotional Com-
mentary on First Peter, 1967; Biblical Predestination, 1969;
Historiography: Secular and Religious, 1971; II Peter: A Short Commen-
tary, 1972; The Johannine Logos, 1972; Three Types of Religious
Philosophy, 1973; First Corinthians: A Contemporary Commentary,
1975; Predestination in the Old Testament, 1979; The Concept of Bib”
Authority, 1979.

o
>

Box 817, Phillipsburg, N.J. 08865

2
:
o>
=
=
ee
ISBN: CO a
—