Danto - Historical Language and Historical Reality

Historical Language and Historical Reality
Author(s): Arthur C. Danto

Reviewed work(s):
Source: The Review of Metaphysics, Vol. 27, No. 2 (Dec., 1973), pp. 219-259
Published by: Philosophy Education Society Inc.
Stable URL: http://www.jstor.org/stable/20126419 .
Accessed: 03/09/2012 01:06
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Philosophy Education Society Inc. is collaborating with JSTOR to digitize, preserve and extend access to The
Review of Metaphysics.
http://www.jstor.org
ARTICLES
HISTORICAL LANGUAGE AND HISTORICAL REALITY

ARTHUR C. DANTO
Metaphysics fragmentary, irresponsible, and half-awake, and uncon

scious that she is metaphysical, spoils two good things when she injects
herself into a natural science.
William James, The Principles of Psychology
JL intend by the end of this paper to have dissolved a number of
problems in the philosophy of history, by showing their structure

to be one they share with a very wide class of philosophical prob
lems which themselves are due to confusing two ways in which
language is related to the world. This to condensing

leads causal
and semantical information into single formulations, and the prob
lem disappears when these are distinguished. As elsewhere in
philosophy, diagnosis shades into therapy, since the description of

the problem is all that is required for its solution. And, again as
always in philosophy, the solution to such philosophical problems

comes from solving the problem of philosophy itself. So I shall
reach the problems which animate this paper only by progressively
reducing the scale of analyses that otherwise are meant to reveal
structures which always are the same. The philosophy of history
is just philosophy writ small.
I
There is a form of intellectual controversy, exhibited through
out the nineteenth century and into our own, which is less accessi
ble because of a radically different order than certain controversies
it appears to resemble, namely those which sprang up dramatically
between science and religion in this era. Those latter controver
sies developed chiefly because it was at first supposed that religion
was in possession of factual truths which entailed answers incom
patible with those offered

by science, to just the same factual ques
tions : the age of the universe, the mutability of species, the com
munity of men and beasts or, for the matter, the historicity of
sacred writ itself. On the basis of criteria supposed appropriate
to the resolution of scientific controversies as such, the factual
220 ARTHUR C. DANTO
claims of religion were one by one infirmed, and it seemed then

a reasonable projection that more or less the same would happen
in any future such collisions. This left certain options open for
religionists. They could, like adherents of any infirmed theory,

simply give them up like good scientists are supposed to do. Or
they could
attempt to shun refutation by regarding religious state
ments as mere mythic paraphrases of recoverable literal scientific
statements which were in fact true. Or religion could constitute
itself a specialized science, concerned with an area of reality the
"other" sciences did not touch upon, purging itself of any state
ments which then could conflict with those of the "other" sciences.
Or finally, it could radically redefine its own scope and office by
foreswearing any claim at all to factuality, so that questions of
truth or falsity simply no longer would apply, and controversies
between it and any discipline to which such questions did apply
would be as inconceivable as a truth-contest between chemistry and
opera. The Wittgensteinian suggestion that it is not the purpose

of religious language to state facts may be seen as a late response
to a dilemma which is then resolved by assigning religious utter
ance to one language game and physics to another. Were this a
suitable disposition of the matter (I believe it not to be), religion
ists might be grateful (as it is the mark of religionists always to
be) for the suffering undergone, for though in the light of it the
dilemma need never have arisen, still, had it not done so, religion
might never have arrived at a proper consciousness of its own
deep character. So the recent history of religion recapitulates its

own favorite episodes of trial and purified renascence.
The superficially parallel controversies I wish to anatomize at
the outset are between science and philosophy, again on what
seems at first the common assumption that philosophy is offering
factual descriptions which collide with those of science : as though
there is no difference to be drawn between philosophy and science
so far as concerns the enterprise of furnishing factual descriptions
of the world. On the other hand, and in contrast with the case of
religion, these controversies appear to be structured by a curious
dialectic. It is not a dialetic of the progressive sort, leading to
higher syntheses, but a static, circular, oscillating dialectic, in
which opposed positions rotate alternately and tediously into
domination over one other. These are controversies which do not
HISTORICAL LANGUAGE AND HISTORICAL REALITY 221
appear open to settlement by Scientific Method, since it is exactly,

as we shall see, Scientific Method which is in issue: Even if the
duelers decide to first duel over the location of the dueling grounds,
they must find some place to settle the preambular dispute, and it
is the regressive character of the dialectic I want to describe that
one cannot begin the fight without presupposing a victory which
is hollow.
Let us consider, for illustrative purposes, the grand contro
versy still being ritually enacted in a major part of the world as

the substance of philosophy, between Idealism and (Scientific)
Materialism. I shall paint the antagonists with the lurid text
book colors through which they perceive one another. Grosso
modo, then, Idealism shall be the theory that the entirety of reality
depends upon thought, and that there can be no intelligible thought
of a thought-independent reality. One way of glossing "depends"
here is Kant's. Kant claimed that just those concepts required in
order for reality to present itself as intelligible are furnished by
theshaping apriori action of the understanding, so that the world
as we know it comes under structures we ourselves supply. At
least aller Erscheinungen, characterized by Kant as blosse Vorstell
ungen so present themselves. Dinge an sich, which Kant super

stitiously retained in his ontology, do not present themselves, and
so do not in this sense "depend" on thought. It was this surviv
ing realism which Hegel smartly excised, not as false but (appeal
ing to a criterion of meaningfulness only a century later to become
explicit) as meaningless, leaving reality in its entirety dependent

(in Kant's sense) upon thought, there being only one reality,
whose structure is now the structure of understanding. As an
influential critic put it, it was Hegel's theory that "The real world
''
is only the external, phenomenal form of the Idea, and hence the
precise inverse of the true theory (his) that ideas are the internal
form of the real world: "The external world as reflected by the
human mind, and translated
into forms of thought." In a deep
sense, this remains a more

paraphrase than a rebuttal of the thesis
it was meant to overturn . . turn side up again"), for
(". right
when we superadd the dialetic to the concept of materialism, we
find, since dialectic was meant exactly to characterize the processes

of intellect?die wissenschaftliche Anwendung der in der Natur des
Denkens liegende Gesetzm?ssigkeit- ? that the laws to which reality
222 ARTHUR C. DANTO
confirms are psychological. And the resultant mentalization of

matter hardly could be more in the Hegelian spirit. The hurt
reply to this would be that at least it is not being claimed that
reality is dependent on thought, and it is to the further glossing
of this that I now turn.
Intellectual history is a tale of vulgarization. When a man
claims that reality is dependent upon mind, one can interpret this,
and men in fact did interpret it as a causal proposition, madly to
the effect that the world in some dark way is extruded by res
cogitans. Against this startling but almost immediately dubious
theory, Marx's riposte seems an almost Johnsonian reflex of com
mon sense responding to an extravagantly false violation of itself.
Brains need not secrete thought, as Cabanis proposed, nor extrete
them, as Vogt suggested in a crass metaphor. But something like

what appears to be factual evidence appears available in support
of the claim that, taken as a causal proposition, the truth indeed
in the inverse of Hegel's view. The world is considerably older
than the earliest brained creatures, and whatever in fact may be
the connection between brains and thoughts, it is likely that think
ing creatures are creatures with brains; and since these evolved
in a world which once was without them, how could they have
caused that world? But secondly, there seems again considerable
evidence that what thoughts we have are capable of explanation
with reference to things which take place outside those thoughts
and often outside the person who has them. Marx, for
example,
who was less a materialist than a
realist, supposed one might
plausibly explain the general way men think in terms of their class
location and relationship to the springs of production: that their
mode of reading the world is an interiorization?a coming to con
sciousness?of the material circumstances of the class to which one
belongs. And though not an epiphenomenalist, since he supposed

that class-consciousness (as opposed to consciousness of one's
class) might enforce theclass-structure, supposed he

that the
circumstances, which explained the existence of classes, were "ma
terial," and antedated any "coming to consciousness." There is,

in brief, a wide class of quite reasonable theories and a body of
confirmed fact which appear massively to disconfirm Idealism.
But of course they do so only by first transforming Idealism into
a science, characterized through its laughably hopeless attempt to
contribute to our understanding of the world. Idealism has, at

this level, nothing to say. It first must transform its rival into a
philosophy. Let us then sketch its reply.
The theories advanced against The Idealist are causal the
one and all. Each applies a scheme, the scheme of causality,
ories,
which hardly can be derived from reality since it enters into the
definition of reality as we know it. The causal concept is part of
what determines the structure of intelligibility for a world which
science, under such a scheme,
may go then on to describe. But it
can hardly call in question what it presupposes as a condition of
its own coherence. Idealism is a theory about (among other
things) causality. The causality it means to understand is just

the sort of causality exemplified in all the theories marshalled in
refutation of it as though it were in fact just another causal
theory ! As a causal theory, it might indeed be wrong. But that
is not what it is, and Materialism may engage with it only if it has
a rival theory of causality to offer, and takes then its responsibil
ities as a philosophy seriously. It is concerned with what its shal
low rival all the time employs, and the Idealist thus wins against
the latter not by descending to the latter's level but by demonstrat
ing that no war between them is possible since there is a level for
it to descend to. Only, then, if Idealism is a science or Materialism
a philosophy can there be an issue between them to be joined. In
the former case, Idealism is false, but in the latter case Mate
rialism cannot refute what it mistook as its rival as though the
matter were a scientific one.
It will be instructive, before attempting to generalize beyond

this, to consider, with equal crudeness, two further seeming con
troversies which have much the same dynamisms. Then we can
turn to history itself.
A. Empiricism as we construe it today is a theory to the effect
that the total set of meaningful descriptive predicates either are
elements of a base vocabulary consisting exclusively of sense
predicates, or are resoluble via avenues of definition or reduction

to definientia composed exclusively of elements in the base vocab
ulary. This is a considerably chastened transform of a once

robust theory to the effect that things either are or are composed
of ideas (= Kant's blosse Vorstellungen). Even in Hume, we
begin with a psychological theory, in wThich ideas are atomic or
224 ARTHUR C. DANTO
molecular, and where the intra-atomic forces

(the psychological
counterpart of gravitational force in particle mechanics) turn out
to be analogous to those semantical forces which hold words to
gether as texts; and we end with a theory of reality in which things

are swarms of ideas. So empiricism has vacillated as between a
prescriptive theory of meaning and a descriptive theory of the

world. I know a philosopher who lost all interest in sense-data
when it began to dawn that things are not literally constructed
out of sense-data in a way rival to and more fundamental than the
way we believe chairs, say, literally to be constructed out of mol

ecules or splinters. Now as a scientific theory, empiricism ap
pears to run counter to certain scientific facts ; e.g., that sensations
are caused by things; e.g., pains by pins, impinging the body's
sensory peripheries. How can things be then made of what they
may be invoked instead to explain? The transforming, immediate
reply (see Berkeley, Price, and Merleau-Ponty) is that all the
causal laws employed in such causal explanations themselves rest
upon the evidence of the senses : laws relating things to sensations

may be formulated as laws relating sensations with sensations.
So, even the science of physiological perception finds itself lodged

in the superstructure of a system whose base
it pretends to call in
question. To then say that the elements in the base may be ex
plained merely turns the circle through another revolution, and the
rotation of the controversy is endless and comical. But of course
the vicious dialectic evaporates when empiricism recognizes that
it is not a theory of reality but instead a theory of meaning.
B. There is a singular influential theory of language acquisi
tion abroad, according to which the discrepancy between language
input and output as linguistic competence must be compensated
for by assigning to each learner an innate structure, a
linguistic
virtual language, upon which each actual language is based. Sup
pose we superadd to this theory some speculative remarks of lin
guistic visionaries likeWhorf or Nietzsche, according to which the

structure of the world we live in is a reflection or projection of a
grammar, so that permutations in grammar would entail permuta

tions of reality, and every manual of grammar is a transvestite
treatise of metaphysics. Combining this with the theory of innate
grammar means, of course, that at the deepest level, we all live in
the same reality, since we share a common deep grammar. Well,
a man might argue that it is difficult to see how that structure of

reality should be one with the structure of language, since different
languages might have evolved, that is, deeply different languages.

Perhaps Martians, for example, have a different deep grammar
than that which palimpsestically bleeds through all of ours. This,
to be sure, would mean an alternative reality for Martians, since
it would imply a different spontaneous metaphysics. This would
not make it not merely a serious question of how communication
with Martians then would be logically possible. It also raises the
question of whether an alternative reality itself is an intelligible
concept. For are we not, after all, describing such possibilities in
?our language ? And so the very notion of alternatives is some
thing the structure of our language makes intelligible? And hence

'c
is internal to our grammar ? And hence our reality" which must
then just be reality since no external alternative to it can intelli
gibly be framed? So that our language is a kind of fatality, dash

ing any hope that
by changing our grammar we might create a
better reality, one more suited, as Nietzsche hoped, to human ful
fillment? In any case, we cannot speak of the limits of our lan
guage save through our language, and so presuppose the struc
tures we want to speak of as limited and relative ?
I have not aimed at precision in sketching these controversies,
but their structure ought to begin by now to emerge into perspicu
ity, and so ought the dynamisms through which they are generated.
I believe in part at least each such controversy?in contrast either
with philosophical controversies or with
ones?arisesscientific
when one side is philosophical and the other scientific, and the
former is disguised as inadequate science or the latter disguised
as incompetent philosophy. Then, to be addressing
appearing
themselves to a common problem, the antagonists circle inter
minably about an illusory fixed center. The controversies of this
order may be dissolved only by defining exactly the boundaries
between philosophy and science. Because mutual self-redefinition
appears to be involved, the disputes in a sense resemble those be
tween religion and science. But whereas I am uncertain what
interest could attach to religious propositions unless they could in
principle conflict with those of science, since religion one would
hope, must pretend to tell us truths about the world (ourselves

included), I am convinced that philosophy never has truths to tell
226 ARTHUR C. DANTO
about the world, and so cannotconflict in principle with science or,

for what it may be worth, with religion either. I want now dog
matically to state where I believe the boundaries are to be drawn.
II
The incommensurabilities between scientific and philosophical

propositions reflect the incommensurability of two distinct rela
tions in which language stands to the world. In one
relationship,
stands to reality merely in the part-whole relationship :
language
it is amongst the things the world contains, and is merely a further
element in the order of reality. In its other relationship, language
stands in an external relationship to reality in its entirety, itself
included when taken as included in the inventory of reality. It is
external primarily when we construe it in its capacity to represent
the world, and hence in its capacity to sustain what I have else
where termed semantical values, for example "true" and "false."
Anything which is representational in this sense I shall construe as
language, e.g., pictures, maps, concepts, ideas, these also sustaining

this double relationship with the world. I should suppose that
there is nothing in the world unsusceptible to true representation
modulo some convention of representational adequacy, and this, of
course, is true of language itself when taken in its intraworldly
location : thus descriptive linguistics. So in a way it is imaginable
that there should be an ideal representation of reality which would
be complete, a perfect mapping of description onto the world, the
entire complex, language again included, recoverably projected
onto language as a representation : the sort of thing dreamt of by
Borges. But then there stands between language so construed and
reality so construed, a metaphorical space which is part neither of
language nor reality, the two separated, as it were, by what the

continental philosopher would designate as un rien. So the inter
space between and the world would . . ce n'est que
language (.
rien) not itself appear in the map. Satisfaction of the sorts of
corresponential connections which include such concepts as truth,
denotation, instantiation, exemplification, since the latter hold be
tween reality and what is mapped onto it, are never part of the
map. It is for such reasons that the Tractatus, which primarily

is addressed (as I would argue that any philosophical work is) to
just these connections, has no room in the language it characterizes
for the propositions of the Tractatus.

This is its magnificent in
sight. Its major failing is to suppose there is some special way

language and reality each must be?respectively composed of
atomic propositions and co-structural atomic facts?in order that
there should be mapping at all; and its assimilation of the vehicles
of description to what is after all only a conspicuous kind of rep
resentational vehicle, namely pictures. But let us pursue our
more general diagnosis.
I shall suppose that science may bethought of as at least an
attempt at describing the world, in adding to the map, as it were,

or in rectifying or even, in scientific revolution, replacing the map,
whatever may be the special strategies worked out for achieving
these ends. Were philosophy like science, it too would be making
maps of reality. It is not that, however, for its province is the
interspace between language and reality, and so to the content of

our representation of reality it contributes nothing. Science and
philosophy then are at right angles, and intersect but cannot pos
sibly conflict, since they lie in logically distinct planes.
The duplex relationship between language and the world, and
especially in those cases where language itself in its intraworldly
relationship is its subject in the extraworldly relationship of de
scription, is exceedingly treacherous, as the entire history of phi
losophy confirms. For in describing language as part of the
world, and as standing in such dubiously worldly relationships to

other parts, we are almost irresistibly inclined to think of all the
relations between language and reality as though they were of the
sorts which relate any part of reality to another : say through the
causal relationship. But then those terms which have reference
to the extra-worldly connections?words like "truth," or "exis
tence," or "representation," or "reality" itself?become remark

when they are treated as though they were ordinary
ably puzzling
or even special descriptive concepts :we look for something they
describe, or treat
them, so to speak, as mere noises. One response
to this is to naturalize these terms, forcing them to do merely de
scriptive work, e.g., as the Pragmatists attempt to collapse the
concept of truth onto the concept of success. Or to treat these

words as are
ordinarily
they used, e.g., describing in the uses
ordinary language of such words as "true," these descriptions

then coming to form part of that map of reality which describes
228 ARTHUR C. DANTO
the way parts of language are used in the world. The latter enter
prise then comes into conflict with those theories of truth which
treat it as having reference to the relationship between language
and the world, e.g., as in correspondence theories :which is why the
famous quarrel between Strawson and Austin seems at once so
futile and so
inconclusive, since one was addressing himself to a
'' ''
description of
true in one plane and the other to an analysis of
the relationship it stands for in another plane. Indeed, that con
troversy is of a piece with those we have attempted to characterize

thus far in this essay, which arise from the attempt to naturalize
a semantical concept, or semanticize a natural one. Whenever
this happens it is reflected in consciousness by a stale dialectical
comedy which can only be ended by unmasking one of the un
wittingly disguised controversialists.

We may illustrate this through the causal theory of percep
tion, in which both sorts of
relationships are involved, since our
percepts are at once considered, under that theory, as within and

without the world. It is because of this that standstills of the sort
exemplified (I) are naturally
under generated. The causal theory
arises primarily for representationalist theories of perception?
where one's perceptual connection to reality is mediated by an
intervening entity, a percept (or Id?e) which then resembles or
represents that reality. This supposes, in "veridical" perception,

that there is an object o which my percept, as it were, refers to
and correctly represents (Descartes and Eussell at various points
suppose that it resembles o, but this is to be attributed to a further
instance of domination by a picture theory of representation).
Failure of representation then defines hallucination as inadequacy
of representation then defines illusion, but what seems obviously
true is that at this level, percepts have been construed on a model
seemingly more suitable to sentences?a representational theory

of language is not at all odd?and veridical perception thus is
analyzed along logical lines suggested by a correspondence theory
of truth. But it is characteristic of philosophical theories of per
that they are shaped on linguistic models : semanticization
ception
of perception is not endemic to the representational theory of per
Now when one semanticizes in this manner, it automat
ception.
ically follows that veridicality is something externally conferred
upon percepts, which themselves then are neutral as to whether
they are veridical or not : again like sentences. Causality is intro

duced to eliminate a class of cases in which a man m enjoys verid
ical perception without having knowledge. Thus, suppose I am
having a set of perceptions which indeed veridically represent the

world, but it is mere coincidence that they do ; e.g., I am caused to
have these percepts by factors having nothing to do with that in
the world which makes my percepts veridical, or which satisfies
what we might term the veridicality-conditions of my percepts.
I might be dreaming or drugged, and the content of my experience
even so might be indistinguishable from what it would have been
had I instead been staring reality in the face. So one adds to the
notion of veridicality this : a man m has a veridical of
perception
an object o if (i) o satisfies the veridicality conditions of the per
cept p and (ii) m is caused by o to have p. Causality closes, or
hopes to close, an epistemic gap in the analysis of perceptual
knowledge. It is no concern of ours at this point whether it suc

ceeds or what other gaps if any the analysis of perceptual knowl
edge opens up. The point is that percepts have been made to
stand in two interestingly distinct relations to reality, one causal
and the other semantical. It is worth stressing that the object o
also becomes doubly related: as the cause and the subject of p.
It is through the causal connection that perception appears to
come under the scope of science, specifically the physiological ex
planations of how we are caused to have the perceptions we have.

And from this
vantage point, there
logical is no skepti room for
cism, unlessthe general skepticism to which causation itself is sub
ject to, in as much as perception and its causes, since equally under
causal laws, are indifferently in the same level of reality : causality
is intra-worldly. It is through the semantical relationship that
skepticism enters, for
percepts our do not bear their veridicality
on their faces, and we may logically undergo all the same experi
ences whether there is a world for these to denote, and whether we
get it right or not. To be sure, causality was meant to help us out

of this difficulty, but percepts do not wear either their causal
credentials on their faces : nothing about a percept, any more in
deed than anything about anything, logically guarantees its own
causal history or even that it has a causal history. So if we think
about representation, there is a problem of the external world,
though if we think about causality there is none since by virtue of
230 ARTHUR C. DANTO
causality we are, as it were, already within reality which is defined
by the causal order. Science is concerned with causality, philos

ophy with representation. That is why philosophical questions
are unintelligible to science and scientific answers irrelevant to
philosophy. If we attempt to collapse one relation onto the other,

there either is no problem or there is no solution. And these re
marks may be extended to cover all of the controversies which have
concerned us, which arise in connection with things which stand,
as language does, internal to a world it is also external to through
the properties of representation.
It is, to conclude these protracted introductory remarks, chiefly
because it is more natural to describe than to think about descrip
tion, that philosophy came, in the nineteenth century and in our
own, to become increasingly a problem for itself. For what did it

describe? Since domain after domain
of reality became the prov
ince of this or that science, it began to seem that philosophy must
describe nothing :
which meant, in one part of Europe, that it must
be meaningless, since meaningfulness was defined in terms of de
scriptivity, and, in another part of Europe, that there must be
nothing and philosophy describes it. Finally, it was thought that

the problem was
generated by descriptivity itself, and
that there
was more to language than description. But this of course was an
unwitting attempt to transform philosophy into a science, con

cerned with the description of language in its intraworldly loca
tion, viz., as a set of gestures (speech acts) and responses. But

this final move left the problem of distinguishing philosophy from
descriptive linguistics, which had already claimed this territory.
But if my sketchy analyses have merit, the space between language
and the world is the habitat of philosophy, and this can never be
part of the domain of science since it always is itself external to
science when the latter is construed as representational of the
world. Philosophy is the study of those semantical forces which
bond language to reality and enable the former to express truths.
It is, to borrow Frege's marvelous characterization of logic, the
science of truth.
Ill
It is, I believe, a contribution to historical understanding to
show how so many of the deepest conflicts which constitute intel
leetual history may be traced to simple confusions I have sought

to sketch : confusions in which, because two sorts of relations have
been inadequately distinguished, philosophy or science bled into
one another. This would be a fine example of the way in which a
non-historical discipline, analytical philosophy/might help history,
in this case, intellectual history, in achieving its descriptive and
explanatory ends. But my announced aim here is with the under
standing of history itself as a descriptive and explanatory enter
prise, and it is my hope that the distinctions I have labored toward

will seriously contribute to this. For whatever is true of language
as such must of course be true of historical language.
By "historical language" I shall have primarily in mind an
open class of sentences which purport when asserted to describe
events which have taken place anterior to their utterance or in
scription. Though my main concern remains with the relations in

which such sentences stand to reality, in this case historical reality,
a few schematic remarks on historical sentences will prove useful.
A. I shall make here no distinction between historical sen
tences and historical beliefs, between those sentences publicly dis
played and those voiced or inscribed within the soul. This is

weakly justifiable in that the sincere assertion of the sentence s by
a man m presupposes that m believes that s ; and the latter in turn
presupposes a higher order disposition on m's part to assert, in a
suitably loose sense of the term, the sentence s. By "suitable
looseness," I mean one can assert the belief that there is a chair
in the room by sitting in it, not merely by such stilted locution as
"Here is a chair." It is strongly justifiable through the fact that
beliefs are sententially qualified, which is to say that there is no
belief which is not the belief that something is the case, and what
one believes to be the case is mapped with sentences. To believe
an historical sentence is to believe that to have happened anterior
to the belief which satisfies the sentence believed to be true.
B. Not being in the perfect or a past tense does not as such
disqualify a sentence as an historical one, so long as the sentence

in question, whatever its tense, entails as one of its truth conditions
a sentence in the past tense. Thus "Johnson ?5 ex-president"
'< " "
entails Johnson was president. George Sand will publish her
third novel tomorrow" entails that George has written at least two
other novels. "Notre Dame du Port is being restored" entails a
232 ARTHUR C. DANTO
sentence to the effect that anterior to its utterance, Notre Dame

du Port sustained deterioration. And so on. Were we to unfold
these grammatically simple propositions into distinct clauses, em
ploying the refractive mechanisms of logical form, we would find

some sentence about the past whose falsehood would confer false
hood upon the whole. So though in the present tense and refer
ring to entities which exist at the time the assertions are made?to
Kennedy's vice-president, to the authoress of Francois le Champ?,

to the dominating Romanesque structure of Clermont-Ferrand?
each of these sentences entails as a condition for its truth some
incontrovertibly historical sentence. We may then disregard sur

face syntax as any criterion for historical language, and look in
stead to semantics.
C. Satisfaction of a truth condition by at least one event ante
rior to its utterance or inscription may be counted a primary rule
of meaning for the historical sentence. As such, of course, this
rule does not discriminate between true and false historical sen
tences, nor should we expect that it should. The distinction be

tween history and fiction is inscrutable from the perspective of
logical form, much in the way in which there is no determining
from the descriptions of a set of possible but not compossible
worlds, which, if any of them, is satisfied by the actual world. One
could not tell by reading alone that the Decline and Fall of the
Roman Empire was not fabricated by Gibbon's creative imagina
tion, and that the saga of the Hobbits was not simply a well-written
chronicle of real events. The Emperor Caracalla is not especially
more credible than Aragorn the King. But one particular feature
of historical sentences must be noted: namely, that the time of
their utterance is one of the conditions for their truth, this being
a matter of logic and not merely of pragmatics. "George Sand
wrote La Mare au Diable" would be false, though somewhere in
time George Sand indeed wrote La Mare au Diable, if the former
were uttered in the wrong temporal relationship to the latter : in
1001 a.D., for example. Since the time of an historical sentence is
a factor which counts towards its truth, historical sentences must
be located in the same time scale as the events they describe. This
certainly distinguishes historical language from other sorts, inas

much as though any utterance has a time, the latter is a truth
condition only for historical sentences. So it is irrelevant for
other sorts of language whose temporality does not penetrate its
meaning. Non-historical language, thus, may be semantically

timeless.
D. It is accordingly analytical to the concept of historical lan
guage that historical sentences be in history, be in definite his
torical relations to the events they describe, if they are true. It
follows from this that historical sentences are external to history
if they are true, since externality has characterized the relation
ship between language and the world when the former is meant
to describe the latter, and hence whenever questions of truth-and
falsity (in contrast with "use") arise. The connections between

time and truth entail that historical sentences are at once within
and without the reality they describe?what I shall speak of as
historical reality?and that this double relationship between his
torical sentences and reality may be deduced, as it were, from the
rules of meaning of historical sentences as such. With so compli
cated a semantics, it is inevitable that these relationships should
have become confused. It is exactly this confusion, as I shall
show, that generates those problems in the philosophy of history

I want to resolve. Perhaps it generates philosophy of history as
such, since it is just such confusions as these which after all give
rise to the pseudoproblems of philosophy, if I may be permitted
an abusive of an earlier analytical generation.
E. It is possible to construe the temporal information implied
through the logical structure of historical sentences merely as a
complex bit of referential apparatus. Thus s as an historical sen

tence uses itself as a reference point to indicate the temporal rela
tionship between itself and the event to which it refers. So we

may exclude, for the moment, asit is in general appropriate to do,
questions of reference from questions of meaning, and concentrate

upon everything other than its temporal relationship to the sen
tence which describes it which is required of an event in order that
it satisfy the truth-conditions of the latter. I write: "George
Sand wrote La Mare au Diable." In the world a
redoubtable,
sometimes transvestite lady consumes a period of time, perhaps
in Nohant, scribbling. These scribblings sum to a novel, La Mare
au Diable. By whatever sartorial mystery words are fitted to the
world, I have produced a bit of historical truth, thanks to a res
olute berrichonne novelist bent on literary fortune. My produc
234 ARTHUR C. DANTO
ing this truth need not be the same as having achieved a bit of
historical knowledge, but our problems for the present do not lie
there. What is crucial is that the temporal deixis we have put to
one side does not, so to speak, penetrate the events it points to,
and so the temporal order as between my sentence and the event
which satisfies it is insofar irrelevant to the former's truth: let us
instead say: bit of truth tout court. And the latter may, by de
liberate suppression of temporal reference, be counted insofar as
timeless. To be sure, there may be temporal information of a sort
which belongs to meaning rather than to reference, and hence
which is satisfied by some temporal features of the event itself,
e.g., that the feat took three years. But this would be time in, not
time of the event relative to the sentence it makes true. When I
restore the referential factor I have found it convenient here to
bracket, I do not, as it were, give further historical information:

at best I communicate that the information is historical. Thus I
do not give you an extra bit of truth in telling you what I have
told you is true. So that it is "historical" in this eviscerated,
referential is no part
sense, of the event in question. Its being
historical only is a complicated way of expressing one of the ways
in which it is related to its description, and indeed, we may say as
much of "historical" so far as it applies to sentences in our own
usage: "historical" is more or less a semantical predicate, like

"true."
F. We may thus dehistoricize historical language by simple

suppression of the relevant
portion of the referential apparatus,
as in the paragraph above: 'George Sand' (timelessly) denotes
George Sand; 'La Mare au Diable' (timelessly) denotes La Mare

'
au Diable; writes' (timelessly) denotes a relationship between the
former and the latter, so that "writes (George Sand, La Mare au
Diable)" timelessly describes an event at Nohant, though writing

takes time and books do not exist before they are written. The
time at which this sentence is satisfied relative to the time at which
the sentence is uttered is irrelevant to its truth, and only becomes
relevant when we rehistoricize. Relative, however, to the fact that
their relationship to the sentences we describe them with are not
part of the events described, we may guardedly characterize these

as "timeless." This is not portentously to record some interest
ing, even astonishing metaphysical property of historical reality.

It is relative to language and then only to one element of the ref

erential liaison between language and reality, that events are time
less :nothing follows regarding the Eternity of the Past. All that
follows is that the pastness of the past is not some essential feature
of the past. So we could in principle give a complete description
of the past?a true description?which would be enriched not at
all by saying that what we had just described was past. All that
would be added would be information about the temporal relation
between sentences and referenda.
IV
Historical language is external to historical reality, but in no

different way than any language is, to which a referential function
is attached. Referenda are not penetrated by the expressions
which single them out for indication, and so are "there," inde
pendent of and unaffected by the fact that they are referred to :
and this is so even if the structure of objects should itself in some
way be assignable to language. That is, there may be a deep truth
to the Whorfian-Nietzschian theory that grammar determines the
way the world is going to be for those who use it, without this in
any way changing the logical fact that reference will remain an
external relationship between bits of language and bits of reality,
even if the kinds of bits referred to may vary from grammar to
grammar. Reference makes realists of us all. Historical refer
ence then is simply reference in a certain temporal direction rela
tive to the referring expression itself: but that the referendum
should be in this respect past does not make it more independent
of and less unaffected by language, than were it instead to have
been contemporary with the referring expression which falls
causelessly upon it. In a way we can say that historical sentences
are true before (= logically prior) they are historically true, in
the sense that all the truth-conditions of a description have to be
satisfied before the question of the temporal relationship between
these and the description arises. Thus the sentence "Vercinge
torix kissed Marie Antoinette" is false, and so stands in no tem
poral relationship to any event. Historical sentences have two

ways of being false, as it were, but the
temporally relevant way is
available only when the sentence in question is ahistorically true.
This may be a clue as to why singular in the future
propositions
236 ARTHUR C. DANTO
tense may, as Aristotle proposed, be reckoned neither true nor

false. "Thereness" is what characterizes reality when situated
at the yonder extreme of a referential line?which always is a kind
of deixis?the hither extreme of which descends from a vehicle of
meaning. So "thereness" is a para-semantical predicate of re
ality (well, "reality" itself is a para-semantical concept), pred

icated as it were of anything insofar as the latter is the target of
a description. But no further characterization of the latter is
implied. In this sense of being there, no distinction may be drawn

between past and present, whose only difference here is a differ
ence in temporal relationship in which they stand relative to a
timed description.
"History professors do not love history because it comes to
pass, but only because it is something that has come to pass." So
thinks Dr. Abel Cornelius, himself a history professor, in Thomas
Mann's marvelous story Disorder and Early Sorrow. Cornelius
specializes in the times of Phillip II, and when his own times, which
are those (evidently not unlike our own, in which social values are
cracking) after World War I, he turns for security and mental

relief to a past he can merely study, away from a present he can
hardly live. He senses, of course, that history is being made, that

he and his children and his servants are being carried into un
known relationships by ungovernable forces: and that the world
he once was comfortable enough in is never going to be the same.
Before this cataclysm, perhaps, his present was very like the past
he takes such pleasure from, where the basis of that pleasure now
is that it ?5 past and is never going to be different. One's only
task then is to find out what happened then, and one is spared the
agony of being part of what, of course, in the living of it?the
Counter-reformation?must have been as wild and precarious and
unstructured as Cornelius' own times (which he of course knows
too). One has the task, it goes without saying, also of finding out
what is happening in the present and not merely living through
what is happening: Cornelius tries to understand. He is inside
and outside his own times at once, as student and as participant.
Logically, of course, and this is the point which concerns me, as

student one is outside, even if as person one is also inside present
events, and one is no more "external," in this respect, to the times

of Phillip II than one's own. It is only that he is excluded from
also being inside the events his professional calling consists in
knowing as much about as he can. He has, for instance, no chil

dren living then, about whom he has to worry. There is little
doubt of course that Abel Cornelius would prefer, as a counter
existentialist, to live in the present as though it were the past, and

to stand only in the external relationship of scholar to his times,
which is the Cartesian posture of a knower set over and against
a logically external world. This posture is available, but not ex
clusively available: because he is a father and a German and a
professor, the external world is what he is in. His children who

live without especially thinking about their times, are utterly in
side the reality from which he is cognitively alienated while being,
through exigencies and distractions, also alienated as cognizor.
Certain philosophers, indeed certain great philosophers
(James and Nietzsche, for example), have felt a certain conflict
between the existential relationship in which we stand to life (we
live it), and the thereness which attaches to any event, past or
present, in virtue of the truth-relationship. They have taken

"thereness" to be some absolute trait of reality in something like
the way people have supposed truth must be an absolute property
of true sentences. They could not, however, accept thereness,
implying fixedness and unalterability, as a trait of reality which
was conspicuously plastic and alterable, available for us to give it
what shape we will, subject only to the most contingent constraints.
They resolved the conflict in a way which makes the difference

between philosophers and historians : by attacking that theory of
truth?the correspondence theory?which appeared to them to en
tail properties of reality repugnant to their experience of the
world. And they invented a new theory of truth, one more con
sonant with their vision of the possibilities of life.
Now the Correspondence Theory of Truth, though it has had
its important errors and perhaps lacks an adequate formulation
even now, is intuitively correct in stating the sort of relation we
want to hold between language and reality when the former de
scribes the latter successfully; and it implies no characterization
of reality beyond whatever characterization is given by the de
scriptions which are true. Their being true is not a further bit of
description, in virtue of which the reality described has a special
property in addition to those it is described as having. And so,
238 ARTHUR C. DANTO
when something satisfies the truth-conditions of a sentence, there

is not some further thing it needs to do to make the sentence true :
being true is not a further truth-condition of the true sentence.
So James and Nietzsche, as well as lesser pragmatists and greater
fascists, confused a parasemantical characterization of reality for
a metaphysical description of reality's deepest trait. The latter
they felt not only inconsistent with the world as they perceived it,
but essentially dangerous in the respect that it induced a sense of
fatalism and hence a sense that there was no way in which we
might change the world (ourselves included, as part of the world),

but could only externally record what was already given and?
there. At once pernicious and wrong, they felt they could only
revolutionize life byrevolutionizing the truth-concept. So truth
had more and more to refer to successful revolutionizing of reality
itself?truth was what worked?and the old external relationship
between and
reality was
impugned : language, like every
language
thing else, had to be part of reality, and those parts of language
to which truth and falsity traditionally attach?declarative sen
tences?were recast as instruments for the transformation and
organization of experience: true if successful in their utilitarian
role, false if failures. This analysis worked best with sentences

whose truth-status is at any rate moot, namely those supposed to
express laws or theories, which now became part of the scientist's
battery of instruments he is charged to design for the generation

of predictions and the smoothing out of experience. It worked
less well for singular propositions, or at least seemed decreasingly
intuitive as the degree of specificity of the sentences to be analyzed
increased, and simply parted company with intuition altogether

when "Christopher Columbus discovered America" proved to be
a sentence about future experience in the archives, with 'Chris
topher Columbus' itself taking on the

status of a
term. theoretical
In general sentences about the past became sentences about the
future. Proper names, which if anything would have seemed par
to play a referential function, were theoreticized
adigmatically
generally ; and indeed the distinction we generally countenance be
tween past, present, and future, were reconstituted, so far as they
were discriminable at all, as under the pressure of a radically em
piricism they had to be, into different compartments for present

and future experiences.
Incongruity with intuition is not an automatically fatal objec

tion against a philosophical analysis, although it would be difficult
to choose between the pragmatist's intuition of a plastic present
and the plain man's intuition about a fixed past. If, in order to be
honest to the former, we must radicalize the truth-concept with the
consequence that we cannot any longer be honest to the latter, it is

not plain that this is a price we must pay. It was after all a late
criticism of the pragmatic theories that they had untoward conse
quences for sentences purportedly about the past, and while we
may admire the pragmatists for accepting and even endorsing

these consequences, since after all it would not do to have a cor
respondence theory of truth for the past and a pragmatistical one

for the present and future, it might have been a better wisdom to
reflect on the question of whether, really, the
correspondence the
ory of truth had such extraordinary metaphysical and human im
plication that it needed rejection in favor of a theory which de

manded so wholesale a redesign of language, truth, and reality.
(Of course it is not in the psychology of philosophy to resist so
grandiose an opportunity.) But I do not think we need merely pit
intuition against intuition to perceive what must be the major ob
jections against the pragmatic theories in general. I can only

afford to mention these here.
The instrumentalist truth-theories were
early members of a
class of theories, the latest members of which have been contrib
uted by (late) Wittgenstein and by Austin, in which sentences are
appreciated as playing roles in the conduct of life, and hence as
being part of the fabric of human existence. Each of these pre

supposes a theory of meaning which essentially is a theory of use,
sentences having different meanings as they have different uses.
Thus if a sentence is used for the anticipation of
experience, that
is its meaning. If it is used to convey encouragement or despair,
that is its meaning. To ask what is its meaning is to ask for what
it is used. And mastering language is something like mastering
some craft, in learning to handle the tools of the trade, using the
right tool for the right job, etc. Instrumentalist truth theories
thus presupposed more or less instrumentalist theories of mean
ing. And one thing which counts heavily in support of such the
ories is that sentences do in fact have uses, even the uses ascribed
to them by philosophers. There is little doubt that sentences
240 ARTHUR C. DANTO
about the past are used for the organization of archival materials.
There is little doubt that moral propositions are used to commend
things or to modify behavior in a certain direction. There is little
doubt again that we use the unit sentence "True." to
express
agreement with some other sentence presupposed reasonably con
tiguous in the conversational atmosphere. And so on. But there,

I believe, the matter ends.
For onething, what we lose in thinking of sentences as instru
ments is any sense of their structure; and with loss of structure
goes loss of explaining two sorts of things : first, those inferences
which are licensed by features of sentences which are simply
blurred when the latter are taken as whole, and secondly, the
obviously crucial way in which we understand sentences by under

standing the relationship of their parts. Were the above story
of linguistic mastery to be believed, mastering language would be
like mastering a ritual, using just the right sentence in just the
right way. in fact, as we now have been
But told, to master lan
guage is to be master of a potential infinitude of sentences, includ
ing sentences never heard or used before. Indeed, these theories

tend to treat sentences as though they were terms, with the finitude
of vocabulary acquisition that this implies and the inductive man
ner in which we learn to use words. Semantical from
theory
Frege through early Wittgensten to Davidson a deep
sees connec
tion between meaning and truth; and while sentences may in addi
tion have uses and social and practical employments of the greatest
and most interesting variety, instrumentalist theories of meaning
lose connection with what impress me as the most fundamental
facts about As we might
language. say by means of an expression
and a concept brought to consciousness by Fregian semantical
analysis: instrumentalism renders everything opaque. We lose

those transparencies which open language up to substitution and
quantification, and by losing the grammatical structures lose, as
well, any possibility of understanding how language works. Of

course, the connection between meaning and truth does not commit
us to any special theory of truth, and the issue of the correspond
ence theory has to be settled anew. Davidson, for example, rejects
it, though mainly by treating sentences, in the Fregian manner, as
names of truth-values, which I do not believe the correspondence
theory at all entails. Whatever the case, since any theory of truth
is going to connect with sentential structure, and since this is

exactly what is lost on pragmatic theories of truth, it is difficult
to see that the connection between meaning and truth can at all be
the way pragmaticists proposed. However finally we are to ap
preciate the relationship of satisfaction which holds between the

truth-conditions (the meaning) of a sentence and their satisfiers,
the relation in question will be of a kind which leaves language
external (even in the case when
to its satisfiers one bit of language
satisfies another, as when it is language about which we speak).
And so we may retain our intuitions about present or past, and
suppose that if the correspondence theory of truth may be exempli

fied indifferently by true sentences about the past and by true sen
tences about the present, the differences between the alleged fixed
ness of the one and the fluidity of the other cannot be explained
with reference to the Correspondence Theory of Truth. We do
not have, conversely to Doctor Cornelius, to live in the past as
though it were the present?we do not have to live the past?in

order to preserve our sense of being in the world. He tried to be
completely external, as the pragmatists thought to be completely
internal, to a world in which we, like our language, stand in both

relations at once.
Historical sentences are partially token reflexive, in that they
are simultaneously used and mentioned?used to describe an event
which is past, relative to the time of the sentence, which thus is
mentioned as a point of temporal orientation. This analysis
works best for true sentences, inasmuch as then there really is an
event earlier in time than the sentence which also truly describes
it. If the sentence is false, as we saw, there is then no event rela
tive to which the describing sentence may stand in any temporal
relation, and this is markedly so when reference fails rather than
predication. For example, "The Battle of Waterloo took place in
Ohio," succeeds in referring to a battle which actually occurred,

although it is badly mislocated (unless the sentence is arch, like
saying the Battle of Waterloo was won on the playing fields of
Eton) ;but "The Battle of Baffinland ..." fails to refer however
we complete the sentence, and either is degenerately false or the
question of truth-or-falsity cannot arise. Failure of reference in

this sense is always the case for singular propositions in the future
tense, at least according to that famous conjecture of Aristotle's:

242 ARTHUR C DANTO
and so, either these are always false or neither false nor true, if
he is right. We may, perhaps, if we acquiesce in the intuitions
regarding time, truth, deliberation, and action which were bruised
by the argument of his opponent in De Interpretatione, work out

an analysis in which we can speak of sentences about future sea
battles or whatever as retroactively true or false, viz., at or after
the time at which the appointed sea-battle occurs or fails to do so,
via a complex proposition referring to a battle, a prediction, and
itself in various temporal relations. But this would be to allow
semantic analysis to be contaminated by metaphysical biases of
various orders, which would be inconsistent in the light of our
criticisms of
pragmatism. Let us merely say that future sen
tences too are token reflexive, and point in the temporal direction
relative to their utterance theyof the event mean to describe.
Then, if for deep reasons there be nothing to point to and the

sentences are accordingly false degenerately, we may retain the
direction of the deixis in our analysis, and keep our ac
temporal
count of timed descriptive utterance consistent. Our concern is
not to say whether future, or any, sentences are true, but the rela
tionship in which they must stand to the world if they are so ; and
this upon no conceptually important character of the
depends
world itself.
In its concern to describe the events to which it refers,
history
?because it aims at truth?may be spoken of as history-as-science.
History-as-science presupposes no special philosophy of history,
apart from whatever philosophy is presupposed by the concept of

truth itself, viz., a specification of the relationship between sen
tences and what satisfies them. Apart from complexities intro
duced by temporal reference, historical sentences stand to histor
ical reality the way any true descriptions stand to any reality it
may be true of. To be sure, as we construe this relationship dif
ferently, we may get perhaps a different philosophy; but this will

be less a philosophy of history than a philosophy of truth, and a
consistent analysis will have to be worked out for history and for
the rest of science, as we saw happen under the pressures exerted
by the pragmatic theories of truth. Historical instrumentalism is
simply a specialization of instrumentalism at large, somewhat less

intuitive perhaps, but nothing greatly turns, as we saw, on that.
There are, of course, special problems of historical knowledge and
of historical investigation : our access to the realities we want to
describe, hence want to describe correctly, is complicated by their

relative pastness. So there may be some special philosophy of
historical inquiry. Still, the strategies for successful inquiry here
will beof a piece with successful strategies for dealing with un
observables generally; and
very while these are interesting cer
tainly, they are less so, I believe, and have anyway been more often
and elegantly discussed than the questions which arise when we
think of history-as-science as part of historical reality. Thus far
we have thought of history-as-science as part of history by dint of
the role it plays in the referential apparatus vis ? vis the events it
describes. But this makes possible the occurrence of exceedingly
complex events, in which beliefs and descriptions about the past
are components, in which, as we might say, men's consciousness of
events becomes part of an event other men later endeavor to be
conscious of. When men's representations of reality become part
of the reality other men seek to represent,begin we to encounter
features peculiar in a way to history. So it is to history as inter
nal to history that I now turn: to history-as-reality, as I shall term
it, in contrast with history-as-science, though of course it is always

the same thing which is at once, because internal and external to
the world, history-as-science and history-as-reality.
Let us begin with a topical matter. In recent times a certain

tension has arisen within the educational conscience of those
charged to teach courses

in general education who themselves have
been trained as specialists. To be sure, this conflict?or contra
diction, as the Marxists would say?has a certain economic basis,

for the activity of teaching general courses is antithetical to the
kind of activity, namely specialized research, which brings aca
demic advancement and glory. But let us attempt to appreciate
the tension less crassly.
The courses, of which the Contemporary Civilization course at
Columbia is the paradigm and perhaps the ancestor, were largely
designed in the twilight of a dilute Hegelian historicism, in which
itwas believed that the understanding of x consisted in knowing
the history of x. So we understand our institutions and practices
by perceiving their emergence from the Dark Ages. It was any
244 ARTHUR C. DANTO
way believed salubrious to think that things have histories : that

morality, for example, should be seen to have an actual genealogy
was believed by Nietzsche to be a morally liberating disclosure.
So the student would be led back to his sources, with the hopeful
result that his perspective would be put in perspective. Even the
perspective itself has a history. To realize that there are other
ways of seeing the world is to wonder if this itself is not just one
of the ways of seeing the world. The student is immersed in a
dissolving relativism which sucks itself into its own dissolution,

and emerges, hopefully, a free spirit. The students of the 1950's,
during which confidence in these courses peaked, were bred to be

relativists. One element, perhaps, in the increasing distaste for
these courses was the desire on their teachers' part to combat this
easy relativism by furnishing some bits of truth. After all, they
did not want the historical truths which gave rise to these rela
tivisms to be swept away by the relativisms they gave rise to. We
now begin dimly to discern as the co-ordinates of these tensions,
the two relations in which historical language stands to historical
reality. We can render its presence more distinct by another

consideration.
Think of those figures whose thought not only is the source of
modern relativisms, but which enters into the fabric of modern life
and must accordingly be studied in those courses which reflect this
thought in their very structure: their thought is (part of) the
structure and substance of these studies. I refer, of course, to
people like Hegel, Marx, Nietzsche, or Freud : those deep but easily
sloganized intelligences. How different, indeed, their thought
often is from the sloganized shadows they cast across the present.
Anyone who has studied closely the conceptual struggles Marx had
with the concept of class or Freud with the concept of sex or
Nietzsche with the concept of power, will appreciate the immense
distances between what they actually said or believed they were
saying, and what men smaller and sillier than they believed them
to have said. And one may generalize upon this decay: the his
torically real x must deviate in some,perhaps in some very con

siderable measure from the historically apparent x. Now among
the teachers of those courses were those whose professional lives
consisted in scrupulous specialized study of the thought of Marx,
Freud, Hegel, Nietzsche, and the like. How, consonantly with the
which define not scholarship but pedagogic re

obligations only
sponsibility can these serious individuals do anything less than set
straight the record and transmit, as they see it, the truth about
Freud, Nietzsche, etc. ? But the truth takes time to tell, especially
when the correspondent realities are such complex architectures
of thought and perspectivity. And on each topic to be covered?
and there are so many and time is so short!?there is a special
truth to tell. The constraints entailed by the Beruf der Wissen
schaft atomize history into an archipelago of islands of truth, each
one fascinating and capable of indefinite exploration: whole life
times would be too short to explore fully some of the major islands
in the chain !
The difficulty with this altogether commendable attitude is
that it, in fact, is the Nietzsche (Freud, Marx, etc.) men have be
lieved to have said what better information reveals them not really
to have said, or to have said with such qualification as to be alto
gether different from the view their global reputation rests upon,
who has exercised the real influence upon contemporary civiliza
tion. The real Nietzsche, perhaps, had no influence at all: or in
fluenced those whose names are writ in water. It is, to borrow
Sartre's impressive distinction, Nietzsche pour autrui who is part
of our history. Nietzsche en soi is just a part, if indeed an inter
esting part, of the past. But it is Nietzsche pour autrui which

vanishes when the past is seen as so many specialized islets of
truth. So just to the degree that we resolve to be true to Nietzsche
en soi are we going to be false
autrui, to pour
and hence false to
contemporary civilization, all

whicha composte is after
of beliefs
about the past, whether true or false. And the point is perfectly
general, after all : it is not what the Commune was, but what it has
come to mean to radical and conservative alike which determines
the political complexion of the present. And what we lose in treat
ing historical reality from the viewpoint of scholarship?of his

tory-as-science?are the forces which relate events to other events,
these being imposed through the beliefs and attitudes of later men.
It is these beliefs and attitudes which make us what we are, so far
as we are part of historical reality. And nothing else can be ex
pected when historical reality is composed in relevant part of the

representations of historical reality on the part of men who live
their lives in terms of these. Only in the special case are these
246 ARTHUR C. DANTO
representations necessarily correct, when the pour autrui of the
past coincides with the past en soi. As we saw, indeed, in connec

tion with historical sentences, an historical belief can stand as such
though false.
This holds for all ages which have had
of history a sense
at all.
Whole eras, after all, go under to be replaced by new ones when
one set of representations of the past surrender before another.
Consider, for example, the Middle Ages from the perspective of the
Enlightenment and from the subsequent perspective of Romanti

cism: the pour-autrui of the Middle Ages was exactly a function
of the differences between the Enlightenment and Romanticism.
But to point out, as good scholars repelled by falsehood, how off
the mark these periods have been relative to the en soi of medieval
times, is in a way to forfeit a proper understanding of the Enlight

enment and Romanticism from ivithin: for these ages were in part
constituted by their reading of the past to be "the Enlightenment
reading-of-the-past." If anything, it would have been believed
the true reading of the past as it actually was. It would be largely
taken by those who held it to have been, in fact, history-as-science.
And indeed, we treat them as historians-as-scientists when we
raise the question of the truth or falsity of their representations.
We can, and as scientists and scholars we ought, to reject the false
beliefs, the myths and superstitutions of ages past. From the
point of view of truth-and-f alsity, wTe are all contemporaries. But

their representations enjoy another life, whether true or false, in
the respect that they are part of what those who held them were,
and hence are part of historical reality. But of course, to treat
representations as part of reality is not to be interested any longer

in questions of truth and falsity. Representations, historical or
any other
kind, are within and without
reality at once. And with
the periods of the past, we can in a way accept this with equanim
ity. But it is not less the case with our representations than with
any. Except that, with these, we feel a certain tensions if we are

scholars ;we want to purge from our representations those which
are false or inaccurate, and replace them with true and accurate
ones. But it is the false or dim ones which, unfortunately, define
our period from within, which form part of our reality. So in a
way, we cannot change our representation without changing our

at the same time. Speaking dramatically, to change our
reality
beliefs about history is in whatever small a degree a revolution of

historical reality. So the scholars who were discontent with false
hood were transforming rather than recording the historical re
ality of the present, and so bore a set of responsibilities distinct

from those which define mere scholarship. In setting forth to
understand the past they, bit by bit, were changing the present.
How disconcerting when our stance outside reality puts us imme
diately within reality. We find, once more, that paradoxically

which has infected historical language from the beginning ! Well,
before pursuing it any further here, let us attempt first to unravel
certain of the problems in the history of philosophy which it was
the announced aim of this essay to achieve. These are all due to
the crossing of externality with internality, a hard thing to avoid
in connection with representations. There will be time after that,
and perhaps some basis, for drawing a moral for historians.
VI
A. Relativism. Logically there is no need that there be any

historical reality which corresponds en soi to a given historical
pour autrui in any given representation of the past. Suppose the

Romans really accepted the account of their origins in the Aeniad,
and believed themselves descended from noble Trojan refugees.
Having an epic past is often a part of having a self-image of
grandeur and esteem?often, but not always?and one may appre

ciate the subtle transvaluations of values which remove us from
antiquity when one appreciates the importance of humble origins

in our mythologies of success and self regard, and thesuspicious
ness which automatically attaches to those who inherited, who did
not have to earn, what they possess. This would be a posture as
mysterious to the Romans as Christianity. Now I have at the

moment no idea of whether the recounted adventures of Aeneus
have any historical weight at all; whether, in short, the book is
more than fiction. But I am prepared to learn, from some arche
ologist inspired by Schliemannian visions, that to each Virgilian

episode there is an objective correlate in historical reality?a cor
relate but no more. The historical Dido may have been a Car
thaginian slut who rented rooms to the shrewd mercenary from

Dalmatia. The point is that whether their beliefs were false or
inflated or accurate and true, we identify what the past was for the
248 ARTHUR C. DANTO
Romans by identifying their beliefs about the past. The past for
them was what it was, whatever history-as-science may ultimately
say about their representations.
There is a striking, famous argument of Russell's that the
world, for all we can tell from the present, could have come
abruptly into existence a bare five minutes ago. We would have
exploded into existence with all the beliefs about the past we in
fact hold, including our beliefs about the age of the world, but all
of these would be as false as the latter. So, logically speaking, we
could live in the perspective of a past which really had no sub
stance. That they are false is after all an external fact about
these beliefs ; their truth or falsity is not part of their content, so
there would be no way of knowing from the beliefs themselves
whether they were true or false. And since the existence of the
beliefs is consistent with either, the truncated historical expanse
of the world playfully proposed by Russell indeed is a logical pos
sibility. It is not, however, a belief we can hold internally, that
our historical beliefs are false. For to have a belief is to believe
the belief in question to be a true one. That is to say, though the
believer may deduce from the concept of belief that his beliefs may
all be false, he cannot pragmatically, in fact, believe regarding his
beliefs that they are false : for the moment one believes them false,
one does not really believe the things in question. Or if one has
them, one doesreally not
believe them false : one is telling oneself
semantical tales. This pragmatic incapacity does not entail, of
course, of any of our beliefs, that they are true, but only that we,
since they are our beliefs, cannot regard their being false as other
than an abstract possibility. In brief, the past for us must be
as the past tout court by us. To believe that Aeneus
regarded
courted Dido, is not just to know what Virgil says about Dido and
Aeneus; it is to hold what he said as true, correspondent to an
episode of dalliance in history-as-reality.
The bearing of this on the position known as historical rela
tivism is the following. It was the theory of relativists that his
tory is just a set of pasts-for-a; ; as though the en soi of the past
were its pour autrui. Questions of truth-or-falsity do not really
arise for pasts-for-#. Pasts-for-# are, after all, just parts of his
torical The past-for-the-Romans an objective
was feature
reality.
of Roman life, as much so as Roman architecture and Roman sew
age, and in just the way in which the past-f or-us is part of contem
porary American reality. But of course to describe the-past-for
us is not automatically to describe any further piece of historical
reality, unless we believe the past-for-us to be, as indeed we do, a

true representation of the past en soi. If all there were for his
torians to do were to report the past-for-them, they would simply
be recording their beliefs about the past. And they could be right
or wrong, accurate or inaccurate, in reporting what their beliefs
were, without the question ever arising of whether these beliefs

were true unless there were another dimension to the past than
the past-for-us. If there were only thepast-for-us, then indeed,
assuming the sort

authority of philosophers sometimes suppose
they have in reporting what their beliefs are, each man would be,
in the protagorian phrase of a noted American relativist, his own
historian. The only errors he would be subject to would be faults
of introspective scanning. But if we really believed that the past
were only the past-for-us, we would have no past-for-us: for the
past-for-us just consists in our beliefs about the past, and these
beliefs like any must be held true by we who hold them. Relativism
may then be true. But neither we nor any relativist who is also
a scientific historian can believe it to be, not so long as either of us
holds any beliefs about the past at all.
To hold an historical belief is to hold that there is (was) some
bit of history-as-reality it describes, external to the belief in ques
tion. And this is so even though the beliefs themselves, in our
case as in that of the Romans, compose a portion of historical re
ality. Historical beliefs are thus internal and external to histor

ical reality, and it was the curious muddle of relativism to have
denied the latter by having discovered the former. "Right you
are if you think you are, ' 'which is the title of a play of Pirandello,
condenses unfairly the relativist attitude, but I have been trying
to emphasize that just when we think we are right only because
we happen to think we are so, at that moment can we no longer
think we are right.
There is, perhaps, a bit more to the matter than just this.
The relativists also noted that it is possible, considered as part of
historical reality, to explain how beliefs came to be held, as nat
urally perhaps, as one explains how certain cities were built. For
getting that their descriptions of these beliefs were themselves

250 ARTHUR C. DANTO
either true or false, it seemed to them that as part of historical

reality, matters of truth or falsity
were as inappropriate to raise
for beliefs as for sewage systems, as though historical beliefs were
only some more historical reality. This brings us to our next

confusion.
B. Historicism. Relativism is but one of a class of philo
sophical and para-philosophical movements in recent intellectual
history which were generated by a naturalistic attitude toward
language and toward thought, where the latter is appreciated in
linguistic or at least in representationalistic terms. Departing

from the incontrovertible insight that language is a natural phe
nomenon in the interesting and relevant respect that its represen
tational properties are subject to causal explanation?the major
insight of psycho-analytical theory is that the content of our
thoughts require and receive explanations in terms of our sexual
histories?naturalisms then go on either to overlook or deny the
external relations between representations and reality in virtue of
which we may speak of representations as true or false. At least
shallow naturalisms do this. Deep naturalisms, to which a gen
uine philosophical interest is attached, attempt to give naturalistic
accounts of such semantical notions as truth and falsity, in such
terms as, for example, the pragmatically favored concepts of suc
cess and failure. Whether deep or shallow, in any case, whatever
naturalisms say about language must automatically apply to the
naturalisms themselves as special ways of representing language
and reality. And here certain dilemmas begin to arise as symp
toms of something having gone wrong.
There is a kind of stale game one can play. One asks if Nat
uralism is true. The Naturalist translates this out to mean : is this
theory successful, does it work. One then repeats the question on

a new level, viz., is it true that it works? The Naturalist then
translates this into the question of whether the naturalistic theory
of truth works. He says it does, or we say it doesn't. The ques
tion is whether either of us is right, and whether our being so can
again be put into the idiom of working or not working ; and like a
drawn chessgame, this discussion endless is an When oscillation.
the continental philosopher says all language is metaphorical, we
slyly ask if that is another bit of metaphor ; and if he says yes, we
say: then your claim is false, because if it is literally true, this

contrasts with metaphoricity, and

paradoxes open like bottomless
abysses ! This is the sort of dilemma I have in mind. For rela

tivism the question arises as to whether it is relative. Historicism,
which insists upon the fact that all theories are historical phenom
ena becomes by its own criterion just another historical phenom
enon. The sociology knowledge of becomes itself just another
social perspective, economic determinism itself economically deter
mined and a reflection of the class locations of its sponsors, and
so on. Each of these theories, and this is true of Naturalisms as
a class, must fall under its own strictures if it is to be general
and true.
Now insofar a given
as naturalism does fall under its own
strictures, it immediately inherits all those limitations it as a the
ory imposes upon those theories which fall under it. But if it fails
to fall under its own strictures, it immediately loses generality,
constituting in itself an exception to all the limitations it wants to
say are inherent in theories as a class : each such
theory holds itself
in hostage. So whether each declares itself an exception to itself,
the way Mannheim appears to have done with the sociology of
knowledge, or else heroically applies to itself, by a scientific exten

sion of the rule of law, whatever limitations it holds are true
of theories as such, naturalisms collapse comically into logically
limited statements.
The dumb oscillations and proto-paradoxes may be blocked,
I believe, by distinguishing the
respects in which theories are
within reality from those respects in which they are outside it.
And indeed, the questions which concern the explanation of the
ories as natural phenomena are altogether independent of those
questions which concern the truth or falsity of theories ; and this

is perfectly general for representations of every sort. Histori
cism, as indeed all these theories, provides us with remarkable
insights into the provenance of human representations. And we
may retain these insights even if the theory which provides them
itself has causal origins and a definite historical location.
History-as-science, indeed, requires that its representations of
history be in history in this sense, and that they have causes. It

does so
largely because it pretends to offer us
knowledge of the
past. Now it is plain that something more is required than that

a representation should be true in order for the latter to be con
252 ARTHUR C. DANTO
sidered knowledge. It is not knowledge unless an explanation is

available as to how it comes to be the case that we have a given
representation of reality, and if it should prove to be the case that
our having a given representation has nothing to do with what
makes the representation true, it is wrong to describe us as having
knowledge. Indeed, to believe myself in the possession of knowl
edge is to believe, in effect, that whatever it is, of which it is knowl

edge enters into the explanation of my possession of it. We may
see this simply in the case of perceptual knowledge. If I believe
on the basis of perceiving rain that it is raining, it would be nat
ural to explain my having this belief with reference to whatever
it is that makes my belief a true one, in this case the rainfall itself.
As we move to more complicated cases, the chains of evidence may
be longer and more tenuous, but the two connections between rep
resentation and the world are if the former is to be con
required
sidered knowledge of the latter : the representation must be ulti
mately explained by whatever it is of which it is true. So it is

knowledge only to the degree that it is within the world under one
connection and without the world under the other.
The application of these necessarily schematic remarks to his
tory is plain. If an historian believes that Louis XVI's attempted

escape to Varrenne was a failure, he presumably holds this is a
reasoned belief, and so has evidence he believes could not be ex
plained unless the belief itself were a true one, and hence that
whatever makes the belief a true one?in this case an event con
sisting in the apprehension of a royal, fleeing family at Varrenne

on June 22, 1791, doomed by revolutionary forces?itself enters
into the explanation of his believing it. Our evidence for the past
are so many bits of the present whose character and existence are
related to those parts of the past which make our state
directly
based on this true. Or at least this is what we
ments, evidence,
are required to believe insofar as we engage at all in the practices
of history-as-science. Hence we are required, as it were, to believe
that historical beliefs are within history if they are without it.
Whether or not this higher order belief, which amounts to the
belief that our lower order beliefs amount to knowledge, is itself
a true one, is a matter for technical epistemology to determine.
Now what relativisms, historicisms, and like doxastic deter
minisms mean to insist upon is that there are further causal fac
tors which color and limit our beliefs about the past or about what
ever, and that these are perhaps inexpungible. It is, however,

wholly consistent with our beliefs constituting genuine knowledge
that they should be colored and limited in these ways, and more
over it is possible that historical statements about these beliefs
should be colored and limited by the circumstances, historical and
other, of those who make them without this disqualifying these
statements as Even science has a history, and all
knowledge.
manner of adventitious circumstances enter into the explanation
of why one concept rather than another should be used at a given
historical period. But on the other hand, we are faced with a cer
tain set of quite special problems, as always, when the representa
tions we offer as true are themselves about the representations of
others :when the beliefs of others are what our beliefs are about.
This brings me to my next set of puzzles.
C. Verstehen. At one point we identified a class of sentences
which, though descriptively applied to objects which are present

relative to the time of their assertion, nevertheless presuppose as
true some sentence about the past. So if the historical condition
entailed by the meaning rules for such sentences fails, then the
sentence in question is false. We can imagine, thus, two indis
cernible objects relative to properties presently observable: two
vases, say, which correspond molecule for molecule. Indeed, let

one of them be a molecular copy of the other, ingeniously produced
by an object-simulator capable of exact reproduction at any chosen
level of fine-grainedness. To the right, then, is a priceless Ming
bowl, to the left its porcelain counterpart, like it in all manifest
respects. Nevertheless, the latter is not a Ming bowl if "Ming"
carries the implications of authenticity and provenance. "Ming
bowl" applied to the righthand object, is a true, and to the left
hand object a false, description, though the objects are in perfect
molecular isomorphism. Parity ofstructures, which is a sym
metrical relationship of course, undetermines the critical distinc
tion between "fine Ming bowl" and "replication of fine Ming
bowl." What is interesting about the example is that the world
might contain objects indiscernible from all now properly called
Ming artifacts and no real Ming artifacts, and no one could tell by
exhaustive examination of these objects that this were so. And
indeed, the possibility may be generalized to cover all entities

254 ARTHUR C. DANTO
whatever: chateaux, vintage wines, dowagers, anything which
might otherwise be described by historical sentences as I have

characterized them. In fact, we can imagine as the extreme pos
sibility the entire world consisting of objects, exactly like the
objects it in fact is believed to consist in, but such that almost
every historical sentence is false. This would be a world in which
things had causal histories wildly different from what we believe

them to have had. Or no causal histories at all if the world, as
Russell proposed in that celebrated conjecture, were to have come
abruptly into existence a brute five minutes ago. Nobody could

tell the difference.
Of course it would make a difference to us. We lose a certain
interest in an object if, though in every possible respect it is like
a Ming bowl, it happens not to be one, but "only" a reproduction.
A woman exactly like my wife, say a woman read off my wife by
a flawless molecule matcher, would still only be a curious sort of
doll, and in any case would not be my wife: I married one of them
and not the other, even though there is no palpable difference be
tween the two. Counting objects the same if they satisfy the same
descriptions, the world might be made up of all the same objects

under non-historical descriptions, but different objects under his
torical descriptions, though none of these differences could strike
even the educated eye. For my purposes, however, it is only nec
essary to note the degree to which our beliefs about the past pen
etrate the language we use even to describe objects contemporary
with those descriptions, the "present world" so called. Russell's
conjecture is incompatible with any ordinary historical statement
applied to the present world, so if we believed for a mad instant

that his conjecture were true, all historical statements would go
false and whole sectors of language would be put out of play. And
in that event objects, however otherwise unaltered, would lose for
us all the interest they have on the basis of the customary histor
ical beliefs. And this is so whether the beliefs are true or false.
Consider in this respect certain stones on the Aventine, which
once were makeweights for the public scales in an ancient Roman
market place. As such they were part of the complex of commer
cial life in Rome, saturated with Zuhandenheit : things merchants
would lay hands on when parcelling out foodstuffs to housewives.
As the empire waned and Christianity rose, churches often and
naturally rose up where marketplaces had been, and those very
stones, having long since lost their Zuhandenheit but felt, perhaps
to have some reason for existence, found their way into odd cor
ners of the churches, as at Santa Susanna. They did not thus
become mere stones, and their to churches suggested to
proximity
the natively susceptible mind of the time a meaning: they were
there as some sort of sacred object?relics, perhaps. In time they
were believed to have figured in the lapidation of saints ! Those
stones are there today, their function in the present being due to
their historical meanings which mere specimens for the
bearing
petrologist blankly lack. What I mean to stress is that the self
identical stones have come under different and non-overlapping
descriptions for differing sets of people who, though they shared
these stones, lived in different worlds. Employing the same logi
cal licence which us to ponder Russell's conjecture, we
permits
might suppose two worlds the objects under all but his
in which
torical descriptions were indiscriminable, but in which all the his
torical beliefs of those who lived amongst these objects differ. In
fact this would require that they have different sets of causal be
liefs as well, since a large and important subset of historical state
ments are those in which a causal presupposition holds : as a cer
tain depression is correctly described as a footprint only on the

assumption that it was caused by a footstep. Not all historical
beliefs are so innocuous as those which focus on religious relics,
and our survival often depends upon our having the right, or at
least not the wrong causal beliefs. But our concerns are logical
rather than and this enables us to suppose a difference
practical,
in worlds as one which does not so much require a difference in
objects but in the beliefs about these objects. It is in this sense

that we live in a different world from they who believed a given
set of stones were created by the Gods of Measures and Balances,
and in a different world again from those who believe their pres
ence to be due to their erstwhile martyrizing impact on early
Christians. To enter another world, as I am now using this ex
pression, would be to see the same objects under different descrip

tions and against the background of differing sets of historical and
causal beliefs. It is here that the much criticized concept of
Verstehen becomes suddenly apt.
The central idea of Verstehen is one which hardly will be con
256 ARTHUR C DANTO
tested save by the most unreconstructed behaviorist. It is that

some reference to the beliefs of agents is required in the explana
tion, hence the understanding of their actions. This is perhaps
dramatized in those cases in which the behavior in question strikes
us as wayward. It would impress a Roman merchant as insane
were he to perceive a man actually worshipping what he casually
weighs potatoes against, as it would impress a medieval Christian
from the Aventine as sacrilegious to see a saintly object thrown
crassly onto the weighing pan by the merchant?like starting a

fire with a piece of the true cross! We defeat the predication of
insanity in the one case and of blasphemy in the other by reference
to differing and in this instance non-overlapping beliefs, though to
be sure, there always is the possibility that the petrolator is a
weird sort of fetishist and the merchant is showing contempt for
superstition, these being descriptions of actions which again re
quire reference to the precise content of these personages' beliefs.

But deviant behavior, as I say, only dramatizes what always is the
case :namely, that it is with reference to their beliefs that we in
part understand and explain differences or similarities in behavior
by human agents generally. It is with reference to their beliefs

that their worlds differ, even when their world contains, under
some suitably neutral description like "stone," the same or even
all the same objects. To be sure, it is less the postulation of expli
cative beliefs in the understanding of behavior which identifies
Verstehen than what has been taken as their peculiar view of how
one discovers what the beliefs are?namely by dint of an act of
empathie intuition: a literal, vicarious occupation, as it were, of

the interior of Other Minds. This has been criticized as pernicious
or gratuitous or impossible, but there are philosophical rather than
merely methodological problems which arise in connection with it
which our distinctions help us, I believe, to appreciate.
There is a certain asymmetry between the way in which we
attempt to explain the actions of others, and the way in which we

attempt to explain our own. With others we indeed invoke their
beliefs as a means to the rationalization of their conduct. In our
own case we make reference instead to the world. The Roman
merchant understands the otherwise weird behavior of the Chris
tian by seeing that if he indeed believes the make-weight to be a
sacred object, then, though the belief itself is false, the behavior
in question makes sense in the light of it, reasoning here being

largely abductive. In his own case, however, his behavior is to be
explained, not with reference to his beliefs but with reference to

what the stones in fact are: plain and simple makeweights. For
our beliefs are in a curious way, perfectly transparent, in the sense
that it is the world rather than our beliefs about it that we think
of when others think of our beliefs. It is this that is epitomized
in the philosophical commonplace that to believe that s is to believe
that s is true. Indeed, when our beliefs become opaque to our
selves, it is virtually as though they were our beliefs no longer;

it is virtually as though we stand to ourselves in the relation in
which we stand to another. So to occupy, as it were, the interior
of another would, by dint of this transparency, not be to have the
interior of another as something of which we would be conscious
or aware :what we instead would be aware of would be the world
as the Other lives it. And we would, in case we achieved the re
quired identification, no longer see it as the world of the other but

as the world tout court: it would be our world, and the beliefs we
sought to empathize with would be our beliefs. But for just the
reasons laid down, when they our beliefs, they are not revealed
are
to us as beliefs at all. To put it with a certain dash of paradox,
we do not occupy our own interiors. We live, rather, naively in
the world. So were Verstehen to succeed, it would fail, for instead
of exchanging, as it were, one psyche for another, one would in
stead exchange one world for another. And the problem of under
standing the Other would remain.

We can put this another way. The beliefs of others are part
of the reality we have to deal with when we explain their conduct
in the world. We can speak of their world meaning only their
beliefs about the world. We cannot in the same sense speak of
"our world." For us, our world is: the world. The reason for
the distinction is plain. Their world is defined by their beliefs.
And to stigmatize them as theirs is ipso facto to refuse to endorse
them :hence to regard them as false. Our beliefs, because we re
gard them as true, are not thought of as our beliefs. And any
attempt to refer to our beliefs is immediately transmitted?to not
our but?to the world. What is curious is that in our own case,
the distance between ourselves and the world which the concept of
truth requires is automatically closed in our own perception of our
258 ARTHUR C. DANTO
situation, because we do not think of the representation of the
world, to which truth properly attached, but to what is repre

sented, namely the world. In our own case we think of ourselves
as within the world when in fact we are external to it, namely in the
respect that we believe our representations true.

It is possible for men to live, as witness the merchant and the
monk, in quite different worlds or to live differently in the same

world. It is the mark of a world that it can be lived, and from this
somewhat vitalistic criterion, one world is as valid as another.
This is the Principle of the Relativity of Worlds. One thinks of
worlds as doxastic environments in which, by mechanisms anal
ogous to those of evolution, men have learned to survive. But we

do not take this Darwinistic view of our own doxastic atmosphere.
It indeed is one we can live. But it is also true. We are by means
of it within and without the world at once.
By this criterion, too, "our past" is just: the past. We speak
of "the past for them" not in the sense of their having different
and special histories, but rather in the sense that their beliefs
about the past are their own. There is no "past for us" but just
the past, and to see ourselves as having a past is just to take the
posture of history-as-science. Verstehen is, after all, understand

ing, not knowledge. Knowledge entails the truth of what is known,
whereas understanding nothingentails so far as concerns truth
or falsity of what is understood. Understanding, however it is
achieved, gives us entry into the world of another in the sense that
it opens up the beliefs of others when these define that world. But
to understand the world of another is not to understand the world,
unless those beliefs are also our beliefs, at which point they become
transparent. Among the things in the world are the ways in which
others live it. And abstractly, our own form of life is also in the
world that way. But because it is our world, we spontaneously
hold it true. And this is what the Principle of the Relativity of
Worlds overlooks.
VII
In the Analytical Philosophy of History, I argued that the
definitive description of the past is not to be given, not because
there are and always will be lacunae in our evidence?which is
banal and contingent?but because earlier events will continue to
receive differing descriptions through the relations in which they
stand to events later in time than themselves. In effect, so far as

the future is open, the past is so as well ; and insofar as we cannot
tell what events will someday be seen as connected with the past,
the past is always going to be differently described. In this paper,
a connected point has been aimed at, this time concerning the
present. For since so many descriptions of present objects and

current happenings presuppose, sometimes as part of their mean
ing, sometimes through what are taken to be sound causal theories,

other descriptions of the past, if the latter are doubtful in any way,
so are the former. There are no doubt descriptions of the present
which are compatible with any account of the past whatever : de
scriptions in an historically or temporally neutral idiom. But for

the rest, I think, it may be said that to the degree that our past is
in doubt, our present?the way we live in the world?is no less in
question. And indeed, our very actions inherit these margins of
incertitude, for what we do can only have the meanings we suppose

it to have if it is located in a history we believe real. If our beliefs
in that history are shattered, our actions lose their point and, in
dramatic cases, our lives their purpose. The present is cleared
of indeterminacy only when history has had its say ;but then, as we
have seen, history never completely has its say. So life is open to
constant re-interpretation and assessment.
It nevertheless remains the ideal of history-as-science to erad
icate the discrepancies between historical reality and history-for
us. To the degree that it succeeds, we live no differently in history
than we do outside history :we live in the light of historical truth.
It is, of course, not altogether plain that truth is to be preferred to
illusion, nor certain that it will make us free. It is only that we
have no choice in the matter once we achieve historical conscious
ness, for we cannot will falsehood or inconsistency. Obviously,
we will live differently in the present as our beliefs about the past
are modified, e.g., to take a current issue, whether ours is a past
of conflict or consensus. It may not be a better or more felicitous
present, but it is not as though we have a choice. For when the
is in doubt, a question mark
past blurs the present, and, since we
cannot will falsehood, our lives persist unclearly until history-as
science has had its say. The present is clear just when the rel
evant past is known.
Columbia University.

Danto - Historical Language and Historical Reality

Uploaded by

Copyright:

Available Formats

Danto - Historical Language and Historical Reality

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Danto - Historical Language and Historical Reality

Uploaded by

Copyright:

Available Formats

Historical Language and Historical Reality

Author(s): Arthur C. Danto

HISTORICAL LANGUAGE AND HISTORICAL REALITY

Metaphysics fragmentary, irresponsible, and half-awake, and uncon

JL intend by the end of this paper to have dissolved a number of

problems in the philosophy of history, by showing their structure

language is related to the world. This to condensing

philosophy, diagnosis shades into therapy, since the description of

always in philosophy, the solution to such philosophical problems

patible with those offered

claims of religion were one by one infirmed, and it seemed then

religionists. They could, like adherents of any infirmed theory,

opera. The Wittgensteinian suggestion that it is not the purpose

deep character. So the recent history of religion recapitulates its

appear open to settlement by Scientific Method, since it is exactly,

versy still being ritually enacted in a major part of the world as

ungen so present themselves. Dinge an sich, which Kant super

explicit) as meaningless, leaving reality in its entirety dependent

sense, this remains a more

find, since dialectic was meant exactly to characterize the processes

confirms are psychological. And the resultant mentalization of

them, as Vogt suggested in a crass metaphor. But something like

belongs. And though not an epiphenomenalist, since he supposed

class) might enforce theclass-structure, supposed he

circumstances, which explained the existence of classes, were "ma

terial," and antedated any "coming to consciousness." There is,

contribute to our understanding of the world. Idealism has, at

things) causality. The causality it means to understand is just

It will be instructive, before attempting to generalize beyond

predicates, or are resoluble via avenues of definition or reduction

ulary. This is a considerably chastened transform of a once

molecular, and where the intra-atomic forces

gether as texts; and we end with a theory of reality in which things

prescriptive theory of meaning and a descriptive theory of the

way we believe chairs, say, literally to be constructed out of mol

upon the evidence of the senses : laws relating things to sensations

So, even the science of physiological perception finds itself lodged

question. To then say that the elements in the base may be ex

guistic visionaries likeWhorf or Nietzsche, according to which the

grammar, so that permutations in grammar would entail permuta

a man might argue that it is difficult to see how that structure of

languages might have evolved, that is, deeply different languages.

thing the structure of our language makes intelligible? And hence

gibly be framed? So that our language is a kind of fatality, dash

hope, must pretend to tell us truths about the world (ourselves

about the world, and so cannotconflict in principle with science or,

The incommensurabilities between scientific and philosophical

Anything which is representational in this sense I shall construe as

language, e.g., pictures, maps, concepts, ideas, these also sustaining

Borges. But then there stands between language so construed and

reality so construed, a metaphorical space which is part neither of

language nor reality, the two separated, as it were, by what the

map. It is for such reasons that the Tractatus, which primarily

for the propositions of the Tractatus.

sight. Its major failing is to suppose there is some special way

attempt at describing the world, in adding to the map, as it were,

maps of reality. It is not that, however, for its province is the

interspace between language and reality, and so to the content of

especially in those cases where language itself in its intraworldly

relationship is its subject in the extraworldly relationship of de