Early Development of Quine Naturalism

Boarding Neurath's Boat: The Early Development of Quine's
Naturalism
Sander Verhaegh
Journal of the History of Philosophy, Volume 55, Number 2, April 2017, pp.
317-342 (Article)
Published by Johns Hopkins University Press

DOI: https://doi.org/10.1353/hph.2017.0031
For additional information about this article

https://muse.jhu.edu/article/654304
[ This content has been declared free to read by the pubisher during the COVID-19 pandemic. ]
Boarding Neurath’s Boat:
The Early Development of
Quine’s Naturalism
SANDER VERHAEGH*
abstract W. V. Quine is arguably the intellectual father of contemporary natural-

ism, the idea that there is no distinctively philosophical perspective on reality. Yet,
even though Quine has always been a science-minded philosopher, he did not adopt
a fully naturalistic perspective until the early 1950s. In this paper, I reconstruct the
genesis of Quine’s ideas on the relation between science and philosophy. Scrutinizing
his unpublished papers and notebooks, I examine Quine’s development in the first
decades of his career. After identifying three commitments supporting his natural-
ism—viz. empiricism, holism, and realism—I piece together the evolution of Quine’s
position by examining the origins of these commitments one by one, showing how
his early views gradually evolved into the mature naturalistic position that would have
such an enormous impact on post-war analytic philosophy.
keywords Quine, naturalism, empiricism, holism, realism, archive studies
1. introduction
next to his celebrated rejection of the analytic-synthetic distinction, Quine’s most
influential contribution to contemporary philosophy is arguably his naturalism, the
idea that “it is within science itself, and not in some prior philosophy, that reality
is to be identified and described.”1 Instead of examining whether our scientific
theories are truly justified and investigating whether the entities posited by those
theories really exist (in some distinctively philosophical way), Quine argues that
we cannot but start from within our scientific web of beliefs, gradually improving
our theories in a piecemeal fashion in the course of our inquiries.
Quine, “Things and Their Place in Theories,” 21.

1
* Sander Verhaegh is Assistant Professor of Philosophy at Tilburg Center for Logic, Ethics, and Phi-
losophy of Science (TiLPS), Tilburg University
Journal of the History of Philosophy, vol. 55, no. 2 (2017) 317–342
[317]
318 journal of the history of philosophy 55:2 april 2017
Many excellent papers have been written about the exact interpretation of
Quine’s naturalism, its scope, and its far-reaching consequences for philosophy.2
Still, little attention has been paid to the genesis of Quine’s ideas on the subject.
Although historians in recent years have contributed significantly to an
understanding of the development of Quine’s views on the analytic-synthetic
distinction,3 not much work has been devoted to the steps Quine took in developing
his naturalism.4
Given that Quine did not endorse a fully naturalistic perspective until the early
1950s, this fact seems particularly surprising. In the early stages of his career,
the 1930s and 1940s, Quine still defended views that he would later dismiss as
first philosophy. Indeed, Quine himself has also noted that he became “more
consciously and explicitly naturalistic” only in the 1950s; that is, “in the ten years
between ‘Two Dogmas’ and Word and Object.”5
The question that arises, therefore, is how Quine arrived at the naturalistic
position that would have such an enormous impact on post-war analytic philosophy.
In this paper, I offer a first attempt to address this question by reconstructing the
steps Quine took in developing his perspective. We can get an excellent overview of
Quine’s evolving views in the 1930s and 1940s from the notes, drafts, and lectures
that are stored at Houghton Library.6 For although Quine’s publications in this
For an outstanding overview, see the papers collected in Dagfinn Føllesdal, Philosophy of Quine,
2
Volume 2 or, more recently, Alan Weir, “Quine’s Naturalism,” and Peter Hylton, “Quine’s Naturalism
Revisited.”
See, for example, Richard Creath, “The Initial Reception of Carnap’s Doctrine of Analyticity,”
3
Paolo Mancosu, “Harvard 1940–1941,” Joel Isaac, “W. V. Quine and the Origins of Analytic Philoso-
phy in the United States,” and Greg Frost-Arnold, “Quine’s Evolution from ‘Carnap’s Disciple’ to the
Author of ‘Two Dogmas.’”
Two notable exceptions are Burton Dreben, “Quine” and Sean Morris, “Quine, Russell, and
4
Naturalism,” who deal with the logical origins of Quine’s naturalism. My account in this paper is to a
large extent complementary to theirs: where Morris, for example, argues that “much of the basic philo-
sophical outlook that led to [Quine’s] later philosophy is present as early as [his] 1932 dissertation”
(Morris, “Quine, Russell, and Naturalism,” 152), I am interested here in the subsequent development
of this basic philosophical outlook.
W. V. Quine, “Two Dogmas in Retrospect,” 398. Although Quine adopted a fully naturalistic
5
view in the early 1950s, he did not use the term ‘naturalism’ until his 1968 John Dewey Lectures,
later published as “Ontological Relativity.” In the lectures, Quine claims to be “bound to Dewey by the
naturalism that dominated his last three decades” (Quine, “Ontological Relativity,” 26). This timeline
is confirmed by the fact that Quine’s seminal “Epistemology Naturalized” from 1969 was initially (in
1965) titled “Stimulus and Meaning” and did not yet contain the term ‘naturalism,’ suggesting that
he had not yet decided to label his philosophy ‘naturalistic’ in 1965. A further clue is that the first
version of “Ontological Relativity” (prepared in March 1967 and presented at Chicago and Yale in
May 1967) does not yet contain the term ‘naturalism’ either (nor a reference to Dewey). Finally, when
Quine, before 1967, speaks about ‘naturalistic arguments’ and ‘naturalists,’ his use of those terms is
distinct from the view with which he would identify the term from the late sixties onwards. See Quine,
“Lectures on David Hume’s Philosophy,” 112, and “Mr. Strawson on Logical Theory,” 149.
See especially the folders titled “Miscellaneous papers” (1925–1931), “Papers in Philosophy”
6
(1930–1931), “Early Jottings on Philosophy of Language” (1937–1944), and “Ontology, Metaphysics,

etc.” (1944–1951). Most items in the first two folders are student papers, whereas the items in the last
two folders are autograph manuscripts, the majority of them related to his planned book on ontology
and semantics. For Quine’s book plans, see Murray G. Murphey, The Development of Quine’s Philosophy,
53. (In transcribing Quine’s autograph notes and letters, I have aimed to minimize editorial interfer-
ence and chosen not to correct ungrammatical shorthand.)
boarding neurath’s boat 319
period were still largely concerned with logic and semantics, the Houghton archives
contain a wealth of unpublished material in which Quine explores more broadly
philosophical topics. I show that, although some features of Quine’s naturalism
were already present in the early 1930s, the wide-scope holism that led him to reject
the analytic-synthetic distinction in “Two Dogmas” also played a crucial role in his
development of the idea that there is no distinct first-philosophical perspective.
This paper is structured as follows. After identifying three interconnected
commitments which support his naturalism, viz. empiricism, holism, and realism
(section 2), I trace the sources of these commitments to three phases within the
first decades of Quine’s career. First, I show that Quine had been attracted to a
behavioristic version of empiricism from the earliest stages of his philosophical
development (section 3). Next, I argue that although some traces of holism were
already present during his graduate years, it took him quite some time before he
started using this idea in answering the question of how an empiricist can provide
a satisfying account of logical and mathematical knowledge (section 5). Finally, I
argue that when Quine first combined his empiricism and holism in the late 1940s,
he only gradually started to grasp the naturalistic consequences of his position,
a process that culminated in the early 1950s, when he first adopted the view that
epistemology is a science, not a distinctively philosophical project (sections 6–8).
2. quinean naturalism decomposed

In order to reconstruct the development of Quine’s naturalism, we first need an
account of what his position precisely amounts to.7 Quine has given many distinct
definitions of naturalism. At the most general level, however, all descriptions of his
position seem to contain two elements: the principled rejection of transcendental
perspectives on reality, and the adoption of a perspective immanent to our scientific
conceptual scheme.8
(NT) No Transcendence: the rejection of any detached science-independent perspective
on reality.
(SI)
Scientific Immanence: the prima facie acceptance of our inherited best scientific
theories and methods.
Applied to epistemology and metaphysics, NT and SI deliver some of Quine’s most

characteristic naturalistic theses. Epistemologically, NT entails that we ought to
abandon “the Cartesian dream of a foundation for scientific certainty firmer than
Note that my aim here is to get clear on how Quine’s naturalism is to be understood. Contempo-
7
rary naturalists defend many different types of naturalism. Still, since Quine’s ideas have significantly
affected most of these contemporary variants, any conclusion about what his ideas precisely amounts
to will arguably be relevant to non-Quinean naturalists as well.
See Quine, “Five Milestones of Empiricism,” 72: “naturalism . . . sees natural science as an inquiry
8
into reality, fallible and corrigible but not answerable to any supra-scientific tribunal, and not in need
of any justification beyond observation and the hypothetico-deductive method”; and Quine “Things
and Their Place in Theories,” 21: “naturalism: the recognition that it is within science itself, and not
in some prior philosophy that reality is to be identified and described.” From the 1980s onwards,
Quine starts using the terms ‘transcendence’ and ‘immanence’ to characterize naturalism. See, for
example, “Responses,” 230: “the immanent is that which makes sense within naturalism, in mediis rebus,
and the transcendent is not.”
scientific method itself,”9 and SI implies that our scientific theories do not require
“any justification beyond observation and the hypothetico-deductive method.”10
Metaphysically, NT shows that the transcendental question of “what reality is really
like . . . is self-stultifying,”11 whereas SI implies that ontological questions are “on
a par with questions of natural science.”12
Although NT and SI seem to be complementary, they are logically independent.
A skeptic, for example, could accept NT and deny SI; she might insist that we are not
justified in accepting our best theories about the world from either a philosophical
or a scientific perspective. Conversely, many present-day non-naturalists will
presumably accept some version of SI, granting that philosophers should at least
start out presupposing that our best scientific theories and methods are largely
correct, yet deny that there is no distinct philosophical perspective from which
those theories and methods might be evaluated.
NT and SI therefore provide us with a first indication of what Quine’s naturalism
amounts to. If we are to reconstruct the way in which Quine developed his position,
however, we need something more. For NT and SI are not just philosophical dogmas
unsupported by any further arguments; if they were, Quine would be vulnerable to
the objection that his naturalism itself is a transcendental, extra-scientific thesis.13
In the remainder of this section, I identify three commitments that support
Quine’s naturalism as specified above: empiricism, holism, and realism. These
commitments, as will become clear, are themselves empirical and revisable theses.
As a result, these commitments should not be viewed as fundamental principles
but as mutually supporting theses, which, as we shall see, are themselves reinforced
by Quine’s naturalism. In the sections 3–8, then, I reconstruct the development
of Quine’s naturalism by tracing these commitments back to their origins in his
work, showing how they have gradually evolved over time in an interconnected
way, eventually resulting in his mature and fully naturalistic position.
2.1. Empiricism
The first commitment supporting Quine’s naturalism as defined by NT and SI is
pretty straightforward. If anything, Quine’s position is backed up by the empiricist
thesis that all our information about the world ultimately comes from sense
experience. Indeed, in “Five Milestones of Empiricism,” Quine presents naturalism
as a distinct stage in the development of empiricist philosophy. At several points in
the past two centuries, Quine argues, empiricism has taken a turn for the better;
and the (for now) final milestone of empiricism is naturalism.14
Pursuit of Truth, 19.

9
“Things and Their Place in Theories,” 21.

10
“Structure and Nature,” 405.

11
“Two Dogmas of Empiricism,” 211. See also Sander Verhaegh, “Blurring Boundaries.”
12
For the objection that naturalism is self-refuting because it is itself not a supported by our best
13
theories, see Robert F. Almeder, Harmless Naturalism, 64, and David Macarthur, “Quinean Naturalism
in Question,” 10.
Again, this is not to say that empiricism is a necessary condition for naturalism. Quine’s three
14
commitments are themselves empirical and revisable theses. To say that any change in these commit-
ments would amount to giving up on naturalism would be unnecessarily essentialistic. Rather, the three
commitments might be better viewed as theses which contribute to the plausibility of naturalism as
defined by NT and SI above. I thank Gary Ebbs and two anonymous referees for stressing this point.
Empiricism bears on Quine’s naturalism in two distinct ways. First, it supports
NT because it rules out many purported extra-scientific sources of knowledge.
Traditionally, many philosophers aimed to ground our scientific theories in an
indubitable a priori foundation. According to Quine, however, the empiricist
can dismiss these attempts as illegitimate; there is no reason to believe that the
rationalist’s self-evident propositions are actually true.15 Empiricism thus supports
NT because it simply dismisses any distinctively philosophical question about the
a priori foundations of science. Secondly, empiricism supports SI in providing us
with an explanation of why we should accept our best scientific theories of the
world. If one agrees with Quine that science is our best attempt to systematically
account for our sensory input, then a commitment to empiricism implies that
one should at least start out one’s inquiries presupposing that our best scientific
theories and methods are largely correct.
Empiricism, however, is not just a philosophical dogma; it is itself supported by
our best scientific theories: “it is a finding of natural science itself, however fallible,
that our information about the world comes only through impacts on our sensory
receptors.”16 Empiricism, for Quine, is simply our best scientific theory about
our sources of knowledge, as is exemplified by the fact that he believes it to be at
least possible that scientists will one day find out that there are other sources of
knowledge as well.17 Of course, the justificatory structure here is somewhat circular:
the respect for science that is embodied in Quine’s naturalism is supported by
empiricism, whereas empiricism itself is a finding of science. It is characteristic
of Quine’s naturalism that he has no problems with such circularity; since there
is no extra-scientific perspective, we cannot but presuppose science in justifying
our prima facie acceptance of science.18
2.2. Holism
In “Five Milestones of Empiricism,” Quine does not only present his naturalism as
a distinct stage in the history of empiricism, he also gives us a glimpse of what he
believes to be the main commitments supporting his position. After defining the
position in the way mentioned above, he distinguishes two sources of naturalism,
the first of which is his evidential holism:
See, for example, Quine, “Lectures on David Hume’s Philosophy,” 54–59. After reconstructing
15
Descartes’s account of self-evidence in mathematics and philosophy, Quine asks: “Why should the self-
evidence of mathematical axioms be a guarantee of their truth, rather than merely a compulsion to
belief—possibly mistaken belief—on our part? And similarly for any other self-evident truth.” According
to Quine, especially the paradoxes show that seemingly self-evident axioms can turn out to be false. See
Quine, “Whitehead and the Rise of Modern Logic.” I thank an anonymous referee for this suggestion.
Quine, Pursuit of Truth, 19.
16
See Quine, “The Sensory Support of Science,” 328: “There is no telepathy, clairvoyance, revela-
17
tion, or extrasensory perception. This is a scientific finding, open, as usual, to reconsideration in the
light of new evidence.”
According to Hylton, “Quine’s Naturalism Revisited,” 50, this way of reasoning is even the most
18
characteristic feature of Quine’s naturalism: “How do we know that the methods and techniques of
natural science are our best source of knowledge about the world? Quine’s predecessors within the
analytic tradition . . . might at this point start . . . invoking philosophical ideas. . . . Quine, by contrast,
insists that the naturalistic claim . . . too must be based on natural science. (If this is circular, he simply
accepts the circularity.) This is the revolutionary step—naturalism self-applied, as it were.”
Naturalism has two sources, both negative. One of them is despair of being able
to define theoretical terms generally in terms of phenomena, even by contextual
definition. A holistic or system-centered attitude should suffice to induce this despair.
(“Five Milestones of Empiricism,” 72)
Evidential holism is the thesis that typical theoretical sentences have no distinctive
empirical content of their own; only clusters of theory are inclusive enough to
imply observable consequences. Whenever we are confronted with an observation
contradicting our best scientific theories, “we are free to choose what statements
to revise and what ones to hold fast” in restoring consistency between theory and
evidence.19 Like empiricism, holism is a thesis that is itself supported by empirical
findings; Quine believes it to be an empirical fact about scientific practice that
scientists have many options to restore a theory’s consistency with observation in
the light of adverse experience.20
So why does Quine believe that evidential holism supports naturalism? What he
seems to have in mind in the above passage is the following: once we realize, on the
basis of holistic considerations, that we cannot translate our theoretical terms into
epistemologically more basic sensory concepts, we ought to acknowledge that the
Cartesian dream of providing an absolute science-independent foundation for our
scientific beliefs ought to be given up. In other words, if Quine’s ideas about the
holistic relation between theory and evidence are correct, the classical empiricist
project of “deducing science from sense data” simply cannot be carried out.21
Actually, Quine’s argument is stronger than this. In an earlier paper, I have
shown that Quine does not only believe that we ought to “despair of being able
to define theoretical terms generally in terms of phenomena,” but also that he
argues that this project is flawed from the beginning, since the sense data to which
the classical empiricists appealed do not constitute a truly science-independent
foundation.22 That is, Quine argued that “[s]ense data are posits too,”23 such that
our ideas about sense experience themselves depend on prior scientific theorizing.
As a result, even if it were possible to translate our theoretical concepts in terms
of sense data, such a reduction would not constitute a truly science-independent
foundation for science:
Various epistemologists, from Descartes to Carnap, . . . sought a foundation for
natural science in . . . the flux of raw sense data. It was as if we might first fashion a
self-sufficient and infallible lore of sense data, innocent of reference to physical things,
and then build a theory of the external world somehow on that finished foundation.
The naturalistic epistemologist dismisses this dream of a prior sense-datum language.
(“Naturalism,” 462)
Quine, “On Empirically Equivalent Systems of the World,” 230.

19
Actually, Roger F. Gibson Jr. identifies three arguments for holism in Quine’s work: one based
20
on scientific practice, one based on language learning, and one based on the problems traditional
philosophers encountered in seeking an acceptable theory of empirical confirmation for individual
statements. Note that at least the first two arguments are empirical. See Gibson, Enlightened Empiri-
cism, 32–42.
“Epistemology Naturalized,” 84.
21
Verhaegh, “Quine’s Argument from Despair.”

22
“On Mental Entities,” 252.

23
The classical empiricists’ talk about a ‘self-sufficient and infallible lore of sense
data,’ in other words, fails to be truly science-independent because our ideas about
our most basic sensory concepts are themselves partially guided by theoretical
considerations.
As a result, the first half of our characterization of Quine’s naturalism—NT
above—is supported by his holism: Quine rejects a detached extra-scientific
perspective on reality because his holistic picture of inquiry leads him to the
conclusion that such a perspective simply cannot be had. Next to his rejection of an
a priori science-independent perspective on the basis of his empiricism, therefore,
Quine also dismisses the possibility of an a posteriori science-independent
perspective. Sense data are simply not science-independent: our ideas about them
themselves depend on scientific theory.
2.3. Realism
Let us turn to the second source of naturalism Quine identifies in “Five Milestones
of Empiricism”:
The other negative source of naturalism is unregenerate realism, the robust state of
mind of the natural scientist who has never felt any qualms beyond the negotiable
uncertainties internal to science. (“Five Milestones of Empiricism,” 72)
Why does Quine cite realism as a source of naturalism? It is my contention that

the answer can be found in “The Pragmatists’ Place in Empiricism,” the paper on
which “Five Milestones of Empiricism” is based. In this paper, Quine compares
his naturalism with the instrumentalist pragmatism of James, Schiller, and Dewey.
According to Quine, these pragmatists “viewed science as a conceptual shorthand
for organizing observations,” such that we cannot ascribe reality to our scientific
posits and theories.24 Now, given his ideas about underdetermination, the view that
there exist alternative theories that would equally fit our observational evidence,
Quine seems prima facie committed to something like instrumentalism as well.
After all, his underdetermination thesis seems to imply that “the systematic
structure of scientific theory . . . is invented rather than discovered, because it is
not uniquely determined by the data.”25 Quine, however, believes that he is not
committed to such a view, precisely because of his commitment to realism.
For naturalistic philosophers such as I . . . physical objects are real, right down to
the most hypothetical of particles, though this recognition of them is subject, like
all science, to correction. I can hold this ontological line of naive and unregenerate
realism, and at the same time I can hail man as largely the author rather than
discoverer of truth. I can hold both lines because scientific truth about physical
objects is still the truth, for all man’s authorship. . . . We are always talking within our
going system when we attribute truth; we cannot talk otherwise. (“The Pragmatists’
Place in Empiricism,” 33–34)
Realism, in other words, is an important feature of Quine’s naturalism; without it,

his position would just be a variant of instrumentalism. After all, instrumentalism
is a variant of empiricism too—a variant, moreover, which is compatible with some
weak varieties of holism.
“The Pragmatists’ Place in Empiricism,” 33.

24

25
A question that remains to be answered is how Quine justifies his realism. We
have seen that both empiricism and holism, the first two sources of naturalism, are
theses which are themselves supported by science. Have we here finally found an
independent philosophical presupposition in Quine’s naturalism? I believe not.
Quine justifies his realism by appealing to semantic considerations. As the last
sentence of the above quote shows, Quine believes that we cannot but think about
our scientific theories as true: “we are always talking within our going system when
we attribute truth; we cannot talk otherwise.” This claim should be taken quite
literally: according to Quine, key philosophical concepts are useless when they
are divorced from their everyday scientific applications. When the instrumentalist
pragmatist accepts science but regards it “as literally false on ontological points,” 26
she presupposes a science-independent notion of ‘truth.’ Similarly, when the
traditional metaphysician asks us about the true nature of reality, she presupposes
that we can separate the term ‘reality’ from its ordinary scientific use. According
to Quine, however, this cannot be done because these very notions are elements
of the conceptual scheme they are supposed to transcend:
We cannot significantly question the reality of the external world, or deny that there
is evidence of external objects in the testimony of our senses; for, to do so is simply
to dissociate the terms ‘reality’ and ‘evidence’ from the very applications which
originally did most to invest those terms with whatever intelligibility they may have
for us. (“The Scope and Language of Science,” 229)
According to Quine, asking whether the objects posited by science are real from a
science-independent perspective is “like asking how long the Nile really is, apart
from parochial matters of miles or meters.”27 Just like our notion of ‘length’ is
useless apart from related notions like ‘mile’ and ‘meter’ and some standard of
measurement, our notions of ‘truth’ and ‘reality’ are useless when one purports
to transcend our conceptual scheme and scientific standards.28
The third source of Quine’s naturalism, his realism, therefore, is supported
by semantic considerations; we cannot ask about the ‘reality’ and ‘truth’ of our
scientific posits and theories in a distinctively philosophical way without stripping
those concepts of their use.29 Quine’s realism therefore contributes significantly
to the justification of both NT and SI. For it provides us with both an additional
reason as to why there is no detached science-independent perspective and an extra
argument for why we are bound to accept our best scientific theories and methods.
26
27
“Structure and Nature,” 405.
See also Verhaegh, Rafts, Boats and Cruise Ships, ch. 3; and Verhaegh, “Blurring Boundaries.”
28
Quine explicitly makes the connection between naturalism, instrumentalism, and the anti-transcen-
dentalist argument in a response to Christopher Hookway: “Hookway finds ‘Two Dogmas’ instrumen-
talist. I think this is fair, and that it applies to my later work as well. But realism peeps through at the
checkpoints, and takes over altogether when we adopt a sternly naturalistic stance and recognize ‘real’
as itself a term within our scientific theory” (see “Responses,” 233).
I take it that this is the reason why Quine describes realism as a negative source of naturalism.
29
Just as holism is a negative source because it entails that we cannot “define theoretical terms generally
in terms of phenomena” (Quine, “Five Milestones of Empiricism,” 72), realism is a negative source
because it implies that we cannot question the reality of the external world intelligibly in a distinctively
philosophical way. I thank an anonymous referee for this suggestion.
3 . e a r ly e m p i r i c i s m a n d b o l d b e h av i o r i s m
Now that we have examined what commitments support Quine’s naturalism as
defined by NT and SI—viz. empiricism, holism, and realism—we can reconstruct
the way in which he developed his position. In the remainder of this paper, I piece
together the gradual evolution of Quine’s naturalism by examining the origins of
his commitments one by one.
Let me start with Quine’s empiricism, the first source of naturalism distinguished
above. Determining the roots of Quine’s empiricism is not a complicated affair.
From the very beginning of his career, Quine was a determined empiricist; nowhere
does he question its plausibility or even take seriously alternative positions.30 In
fact, on the few occasions where he looked back on his intellectual development,
Quine suggested that he was even committed to a strictly behaviorist variant of
empiricism from the very start. Reflecting on his dismissal of intensional notions
in Whitehead and Russell’s Principia Mathematica during his final year at Oberlin
College, for instance, Quine notes:
The distrust of mentalistic semantics that found expression in “Two Dogmas” is thus
detectable as far back as my senior year in college. Even earlier I had taken kindly
to John B. Watson’s Psychology from the Standpoint of a Behaviorist, which Raymond
Stetson had assigned to us in his psychology class. Nor do I recall that it shocked any
preconceptions. It chimed in with my predilections.31 (“Two Dogmas in Retrospect,”
390)
Exemplary of Quine’s early empiricist commitment are his ideas about the relation
between empiricism and pragmatism. Prima facie, pragmatist philosophers have
played an important role in Quine’s early development: William James’ Pragmatism
was one of the first philosophical books he read,32 one of his teachers during his
graduate studies at Harvard was C. I. Lewis, and the young Quine reviewed three
volumes of Peirce’s collected papers in the early 1930s. Still, Quine has never really
understood what it means to be a pragmatist, except if one classifies it as part of
a global empiricist movement:
It is hard to say what constitutes pragmatism. If one considers it a branch of the
empiricist tradition then yes, it is very important to me. . . . But I don’t think that
the influence on me was distinctively American; it was rather one of international
empiricism.33 (“Twentieth-Century Logic,” 60–61)
For a discussion of the way in which Quine’s student papers already advertise a distinctively
30
empiricist picture of inquiry, see Isaac, “W. V. Quine and the Origins of Analytic Philosophy in the
United States,” and Robert Sinclair, “Quine and Conceptual Pragmatism.”
See also Quine, The Time of My Life, 59, and “Autobiography of W. V. Quine,” 7. Although  Quine
31
mentions Watson as an influence, his behaviorism should not be conflated with the philosophy of
science that dominated mid-twentieth-century psychology. In the words of Quine’s critic Jerrold Katz:
“The behaviorism [Quine] has in mind . . . is not the dreaded reductive doctrine of days gone by, but
merely a way of putting the study of language on a par with other sciences by requiring the linguist’s
theoretical constructions to be justified on the basis of objective evidence in the form of overt behavior
of speakers. . . . Quine’s behaviorism is thus a behaviorism one can live with.” See Katz, “The Refuta-
tion of Indeterminacy,” 179–80, but also Hylton, Quine, sect. 4.4, and Føllesdal, “Developments in
Quine’s Behaviorism.”
Quine, “Autobiography of W. V. Quine,” 6.
32
See also Quine, “Replies to Professor Riska’s Eight Questions,” 213. Moreover, it should be noted
33
that in his John Dewey Lectures, Quine praises Dewey not for his pragmatism, but for his insight that
Further evidence for Quine’s early commitment to a strictly behaviorist variant of
empiricism is his approach to analyticity in the 1930s and 1940s. Although Quine
gave up on the analytic-synthetic distinction only in the late 1940s, as we shall
see in section 5, he was already seeking a behavioristically acceptable notion of
analyticity in the early stages of his career. Even in his 1934 “Lectures on Carnap,”
for instance—lectures he would later describe as “abjectly sequacious”34—Quine
proposes that we render
only such sentences analytic as we shall be most reluctant to revise when the demand
arises for revision in one quarter or another. These include all the truths of logic
and mathematics; we plan to stick to these in any case, and to make any revisions
elsewhere. (“Lectures on Carnap,” 63)
Where analyticity had always served an epistemic function for Carnap (at least
in Quine’s eyes)—explaining why our logical and mathematical statements are
justified—Quine here interprets the concept in strictly psychological terms; we
call the truths of logic and mathematics analytic because it is a psychological fact
that we will not give them up in light of adverse experience.35 In other words,
even though Quine here still believes that a distinction between the analytic and
the synthetic can be drawn, he already interprets the distinction behavioristically,
as one between sentences which are and sentences which are not candidates for
revision when confronted with recalcitrant experience.
4. quine’s naturalism in the early 1940s

So Quine was a committed empiricist from the very start. Does this imply that
he had also adopted a broadly naturalistic perspective? Before we move on and
examine the development of his holism and realism, this section explores Quine’s
early ideas about the relation between science and philosophy. I show that his views
were already strikingly naturalistic, albeit in a somewhat embryonic form. Some
crucial elements, I argue, were still lacking—elements which, as we shall see, came
to full development in the late 1940s and early 1950s.
For our present purposes, one of the most interesting manuscripts stored at
Houghton Library is a series of notes from 1944, in which Quine reflects on the
relation between the philosopher’s and the scientist’s tasks in ontology. In the
notes, fragments of which would later be used in both “On What There Is” and
Word and Object, Quine argues for the position that the question “what is there?”
is “broad enough to allow both philosopher[s] and scientist[s] to move about in
it without treading on each other’s toes.”36 He writes:
one should study knowledge, mind, and meaning “in the same empirical spirit that animates natural
science” (Quine, “Ontological Relativity,” 26, my emphasis). See also Heikki J. Koskinen and Sami
Pihlström, “Quine and Pragmatism,” sect. 1, and Peter Godfrey-Smith, “Quine and Pragmatism.”  
“Two Dogmas in Retrospect,” 391.
34
See also, for example, Creath, “The Initial Reception of Carnap’s Doctrine,” 485–86, and Hylton,
35
“The Defensible Province of Philosophy.” I follow Creath in describing Quine’s characterization as

‘psychological’ because, technically, he does not mention behaviorism in his lectures. Still, this is what
underlies Quine’s characterization. In “Truth by Convention,” a paper that largely resembles his first
lecture on Carnap, Quine writes that the apparent contrast between a priori and a posteriori truths
(and thus the analytic and the synthetic) retains reality “behavioristically . . . as a contrast between more
and less firmly accepted sentences” (“Truth by Convention, 102, my emphasis).
“Ontology Notes,” November 5, 1944, my transcription.
36
The philosopher’s task differs from that of the natural scientist or mathematician
no less conspicuously than the tasks of these latter two differ from each other. The
natural scientist and the mathematician both operate within an antecedently accepted
conceptual scheme but their methods differ. . . . The philosopher, finally, unlike
these others, focuses his scrutiny on the conceptual scheme itself. Here is the task
of making things explicit that had been tacit, and precise that had been vague; of
uncovering and resolving the paradoxes, smoothing out the kinks, lopping off the
vestigial growths, clearing the ontological slums. (“Ontology Notes,” November 5,
1944, my transcription)
Given the strong resemblance to his later ideas,37 Quine’s perspective on the
philosopher’s task here is already well developed. The same conclusion can be
drawn with respect to Quine’s ideas about the philosophers’ vantage point:
It is understandable, then, that the philosopher should seek points outside the world
that imprisons natural scientist[s] and mathematician[s]. He would make himself
independent of the conceptual scheme which it is his task to study and revise. “Give
me πoυ στω” Archimedes said, “and I will move the world.” However there is no such
cosmic exile. The philosopher cannot study and revise the fundamental conceptual
scheme of science and common sense, without having meanwhile some conceptual
scheme, whether the same or another no less in need of philosophical scrutiny, in
which to work.38 The philosopher is in the position rather, as Neurath says, “of a
mariner who must rebuild his ship on the open sea.” (“Ontology Notes,” November
5, 1944, my transcription)
Again, the similarity with his later position is remarkable; like the fully naturalistic
Quine, the early Quine is entirely committed to offering a completely science-
immanent perspective.39
A question that naturally arises in the light of these strong similarities, however,
is why Quine never expressed these naturalistic inclinations in his publications, i.e.
why it took him almost a decade before he was willing to publicly commit himself
to a truly science-immanent philosophy. I think the answer to this question can
be found in a different set of notes and drafts. They reveal that, although Quine
wants to commit himself to a fully naturalistic position, he cannot do this because
Cf. Word and Object, sect.56: “What distinguishes between the ontological philosopher’s concern
37
and all this is only breadth of categories . . . it is scrutiny of this uncritical acceptance of the realm
of physical objects itself, or of classes, etc., that devolves upon ontology. Here is the task of making
explicit what had been tacit, and precise what had been vague; of exposing and resolving paradoxes,
smoothing kinks, lopping off vestigial growths, clearing ontological slums.”
Again, cf. Word and Object, sect. 56: “The philosopher’s task differs from the others’, then, in
38
detail; but in no such drastic way as those suppose who imagine for the philosopher a vantage point
outside the conceptual scheme that he takes in charge. There is no such cosmic exile. He cannot study
and revise the fundamental conceptual scheme of science and common sense without having some
conceptual scheme, whether the same or another no less in need of philosophical scrutiny, in which
to work.” Also interesting in this respect is a note from November 28, 1941, in which Quine writes
that he, in his “tentative ontology,” is “[s]tarting at the middle” (my transcription). This anticipates
the first section of Word and Object, where Quine claims that however we “[a]nalyze theory-building . . .
we all must start in the middle.”
At best, one can detect a difference in emphasis. Where the fully naturalistic Quine is prone to
39
focus on the continuity between science and philosophy, Quine in these early notes is more inclined
to emphasize their distinctness. For although they are both working immanently, the scientist and the
philosopher do not yet seem to be concerned with the same project. Rather, as we have seen, Quine
believes that the ontological question is “broad enough to allow both philosopher[s] and scientist[s]
to move about in it without treading on each other’s toes.” 
he has not yet succeeded in combining his naturalized conception of ontology
with a plausibly naturalized conception of epistemology. To see this, consider the
following three fragments:
Here is a straightforward view, likely to be held by a physicist unspoiled by philosophy.
The physicist—even he—is not likely to say his atoms are more real than the tables,
chairs, etc. . . . If physicists do not make the atoms more real than macroscopic
objects, some physicists—tainted with philosophy—do make them less so. . . . But
the macroscopic objects are rather arbitrary as a basic reality, for certainly these are
inferred from a yet more immediate zone in much the way that atoms were inferred
from the macroscopic objects. . . . In latter event we seem to have swung to complete
contradiction of the point of view initially considered. Things are made up now not
of atoms but of perceptions. Seemingly two rival theories of things, the atomic theory
and the sensory theory. Materialism vs. empiricism. Realism vs. idealism. (“Things,”
January 30, 1943, my transcription)
Purpose of the book is to . . . dissociate ontology from epistemology so completely
as to render it immune to the idealist (subjectivist) arguments. (“What it Means to
Be,” March 19, 1944d, my transcription)
There is a sense in which physics might be said to be concerned with explaining the
nature of reality. And who contests this? Primarily the Idealist. . . . The Idealist would
take the perceptions etc. rather as the basic reality, and derive things as constructions,
logical constructs (Russell). The study of how to make these constructions is
Epistemology. And things are composed not of atoms but of perceptions, sense qualia
etc. (“Sign and Object,” October 4, 1944c, my transcription)
In these fragments, Quine is worried about the objections of phenomenalist

epistemology—objections which lead to the conclusion that his scientific
ontology is somehow unreal after all. Where the fully naturalistic Quine replaces
epistemological talk about sense data with its scientific analogue—the stimulation
of sensory receptors—Quine at this point does not yet have this solution at his
disposal.40 Quine’s problem, in other words, is that even though he has succeeded
in naturalizing metaphysics, showing how the philosopher does not require a
transcendental perspective in ontology, he has not yet found a way to naturalize
epistemology, i.e. to get rid of the phenomenalists’ transcendental perspective.41
To be sure, Quine does try to find ways to dismiss phenomenalism. In different fragments, Quine
40
proposes different solutions to the phenomenalists’ objections. One of the most interesting solutions
from a contemporary perspective is the following: “Ontology & epistemology: how they are distinct &
how they are mutually inclusive. Ontology is realistic, epistemology idealistic; but no contradiction”
(“Ontology and Epistemology,” March 27, 1944, my transcription). This mutual inclusivity may remind
us of Quine’s later idea that epistemology and ontology are reciprocally contained (“Epistemology
Naturalized,” 83) although this fragment is of course far too sketchy to attribute such a complex idea
to Quine here. In one passage, Quine comes remarkably close to his later solution: “Epistemology as
a segment of a psychological study. Problem of priority. Answer in child psychology?” (“A Tentative
Ontology,” November 1941, my transcription).
For a different account of Quine’s ideas in the early 1940s, see Murphey, The Development of
41
Quine’s Philosophy, 54–55, who seems to argue that Quine himself was a phenomenalist at this point. I
think it is clear from the above passages, however, that Quine wants to get rid of the phenomenalists’
objections, even if he does not see a satisfying way of doing so. See also the following fragment, where
Quine, after introducing the phenomenalists’ perspective, says: “Bear with me, dissenting reader; I am
going to end up in agreement with you. But first let us see how this thing runs” (“Things,” January 30,
1943, my transcription) and: “we aren’t throwing out philosophy with epistemology, leaving ourselves
It is for this reason, I assume, that Quine, in “On What There Is,” settles for a
pluralistic solution:
the question what ontology to adopt still stands open, and the obvious counsel
is tolerance and an experimental spirit. Let us by all means see how much of the
physicalistic conceptual scheme can be reduced to a phenomenalistic one; still
physics also naturally demands pursuing, irreducible in toto though it be. . . . From
among the various conceptual schemes best suited to these various pursuits, one—
the phenomenalistic—claims epistemological priority. . . . This point of view is one
among various, corresponding to one among various interests and purposes. (“On
What There Is,” 19)
In what follows, I shall argue that one of the crucial steps Quine had to take in
integrating these different conceptual schemes into one single science-immanent
perspective was to develop a thoroughly holistic conception of inquiry.
5. narrow and wide holism

So let us look at the development of Quine’s holism. Reading Quine’s work on the
nature of scientific inquiry from the 1930s, one might get the impression that he
was already committed to holism in the early stages of his career. In his graduate
school paper, “Concepts and Working Hypotheses,” for instance, Quine advances
a view that seems pretty close to the holistic picture sketched in the last section
of “Two Dogmas”:
If a recalcitrant item of experience, belonging to the field in question, should
subsequently arise, modification somewhere in the system must take place, for
it has been noted that a satisfactory conceptual system must accommodate every
experience falling within the field. Thus it is that only the working hypothesis can
stand which has endured without the emergence of any anomaly in the whole mass
of experience since its inauguration. . . . [O]ne has a certain latitude as to where he
may make his readjustments in the event of an experience recalcitrant to his system;
and correspondingly there is some subjective option as to whether a chosen concept
or a working hypothesis is to be branded as the point of ‘error’ in the antecedent
system.42 (“Concepts and Working Hypotheses”)
Still, it would be a mistake to ascribe to the early Quine the radical holism he
advanced in the 1950s, the crucial difference being that he did not yet apply his
holism to logic and mathematics.43 Quine’s holism was still of a narrow scope,
with nothing but physics. There remains ontological problem of essentially philosophical character,
though not epistemological” (“Sign and Object,” October 4, 1944, my transcription). Murphey’s reading
of Quine here might be caused by his being unaware of the above passages in which Quine expresses
his commitment to a science-immanent perspective.
See also Isaac, “W. V. Quine and the Origins of Analytic Philosophy in the United States,” 212–20,
42
and Sinclair “Quine and Conceptual Pragmatism,” 342. Quine’s way of thinking here can, to some
extent, be traced back even to 1927, when he wrote that “[m]an uses as an outline for his knowledge
the natural relationship of all things, so far as he has been able to determine that relationship in the in-
completeness of his data. This web—to change the metaphor—which he has thus succeeded in partially
spinning, he reinforces with synthetic thread of his own manufacture: to wit, the conventional classifica-
tions and man-made systems of compilation which form so large a part of human knowledge. These two
kinds of relationship—the natural and the artificial—work together in such a way that often they are
not to be distinguished one from the other” (“On the Organization of Knowledge,” March 10, 1927).
See also Hylton, “The Defensible Province of Philosophy,” 269–70.
43
applying only to the empirical sciences. Quine had better hopes of explaining
the supposedly a priori character of logical and mathematical knowledge, like
Carnap, in terms of analyticity.44 In “Truth by Convention,” for instance, Quine
still embraces an analyticity-based account of logical and mathematical knowledge:
There are statements which we choose to surrender last, if at all, in the course of
revamping our sciences in the face of new discoveries; and among these there are some
which we will not surrender at all, so basic are they to our whole conceptual scheme.
Among the latter are to be counted the so-called truths of logic and mathematics.
. . . Now since these statements are destined to be maintained independently of our
observations of the world, we may as well make use here of our technique of conventional
truth assignment and thereby forestall awkward metaphysical questions as to our
a priori insight into necessary truths. (“Truth by Convention,” 102, my emphasis)
Although Quine here, as a committed empiricist, clearly does not want to invoke
a metaphysical explanation of our supposedly a priori knowledge of logical and
mathematical truths, neither does he expand his holism to logic and mathematics
so as to claim that our knowledge of those truths is ultimately a posteriori; he still
believes that our logical and mathematical truths are “maintained independently
of our observations of the world.”45
So the question that naturally arises is, when did Quine come to accept a wide-
scope holism that applies to logic and mathematics? There has been considerable
debate about this question in the literature. Richard Creath argues that “[i]t was
not until 1947, and then in private correspondence, that Quine came fully and
finally to reject Carnap’s doctrine that there are analytic truths,”46 whereas Paolo
Mancosu points to a letter Quine wrote to J. H. Woodger in 1942, in which he
argues that Carnap’s “professedly fundamental cleavage between the analytic and
the synthetic is an empty phrase.”47 Greg Frost-Arnold defends an intermediate
position. He argues that, although Quine gave up on Carnap’s semantic version of
Indeed, Carnap himself also combined his analyticity-based explanation of mathematical

44
knowledge with a narrow-scope holism concerning the physical domain: “it is, in general, impossible
to test even a single hypothetical sentence . . . the test applies, at bottom, not to a single hypothesis but to
the whole system of physics as a system of hypotheses” (Rudolf Carnap, The Logical Syntax of Language, 318).
This is not to say, of course, that Quine had not yet taken the possibility of an a posteriori expla-
45
nation of logical truth into consideration. Carnap’s notes of his discussion with Quine about The Logical
Syntax of Language in Prague in 1933, first published in Neil Tennant, “Carnap and Quine,” reveal
that Quine had already questioned Carnap’s strict distinction between the analytic and the synthetic:
He says after some reading of my “Syntax” MS: 

1. Is there a difference in principle between logical axioms and empirical sentences? He thinks not.
Perhaps I seek a distinction just for its utility, but it seems he is right: gradual difference:
they are the sentences we want to hold fast (“Two Dogmas in Retrospect,” 391, translation
by Quine).
In any case, even if Quine had already considered the possibility of gradualizing the analytic-
synthetic distinction in the 1930s, he did not fully reject explanations in terms of analyticity until the
late 1940s, as we shall see. The reason is probably that Quine had not yet fully worked out how an
empiricist could account for logical and mathematical truths without a strict analytic-synthetic distinc-
tion. Indeed, looking back on this period, Quine has remarked that he had “no suggestion of a bright
replacement” when he first criticized Carnap’s analyticity-based account in “Truth by Convention” (“Two
Dogmas Retrospect,” 393). I thank an anonymous referee for drawing my attention to this last passage.
Creath, Dear Carnap, Dear Van, 31.
46
Mancosu, “Harvard 1940–1941,” 331.

47
the analytic-synthetic distinction from the early 1940s onwards, he “was not yet
willing to commit himself to the radical view of ‘Two Dogmas’ until shortly before
writing that piece.”48
Whatever the exact timeline of his adoption of a wide-scope holism, however,
two series of events seem to have been particularly important for Quine’s evolving
ideas on the matter. First, in the academic year of 1940–41, Quine regularly met
up with Carnap and Tarski to discuss, among other things, Carnap’s forthcoming
Introduction to Semantics.49 As Mancosu points out, Tarski at the time defended a
view that comes close to Quine’s wide-scope holism in “Two Dogmas.”50 Already
in 1930, as a note in Carnap’s diary shows, Tarski held that “between tautological
and empirical statements there is only a mere gradual and subjective distinction.”51
Even more revealing evidence that Tarski already defended something close to
wide-scope holism years before Quine came to accept the view, is a letter Tarski
sent to Morton White in 1944:
we reject certain hypotheses or scientific theories if we notice either their inner
inconsistency, or their disagreement with experience, or rather with individual
statements obtained as results of certain experiences. No such experience can logically
compel us to reject the theory: too many additional hypotheses . . . are always involved.
. . . Axioms of logic are of so general a nature that they are rarely affected by such
experiences in special domains. However, I don’t see here any difference ‘of principle’;
I can imagine that certain new experiences of a very fundamental nature may make
us inclined to change just some axioms of logic. And certain new developments in
quantum mechanics seem clearly to indicate this possibility. (White and Tarski, “A
Philosophical Letter of Alfred Tarski,” 31)
Looking back on his 1940–41 discussions with Tarski and Carnap, Quine recalls
how he and Tarski argued “persistently with Carnap over his appeal to analyticity”
in the opening pages of Introduction to Semantics.52 Given the evidence available,
I would suggest that it is reasonable to conclude that Quine must have learned
about Tarski’s wide-scope holism at some point in this period as well.53
Frost-Arnold, “Quine’s Evolution from ‘Carnap’s Disciple’ to the Author of ‘Two Dogmas,’” sect.
48
5. Isaac defends still another hypothesis. According to him, Quine’s motives for not publicly attacking
the analytic-synthetic distinction until the early 1950s were largely political: “Up to the late 1940s,
[Quine] had been content to mute his disquiet for the sake of presenting a united front on logical
empiricism to the American academy” (Isaac, “Missing Links,” 274). Given Quine’s public discontent
with Carnap’s semantic turn as well as any notion of analyticity that could not be explicated in terms
of behavioral dispositions, however, I do not think that these political reasons can explain his refusal
to reject the analytic-synthetic distinction in the early 1940s. Furthermore, also in private Quine was
still actively seeking a behavioristically acceptable definition of synonymy at the time, as is exemplified
by “Foundations of a Linguistic Theory of Meaning,” an unpublished manuscript from August 1943.
In this paper, Quine attempts to formulate an empirically satisfying definition of synonymy, but fails to
find one that lives up to his behavioristic standards. See Murphey, The Development of Quine’s Philosophy,
51–53. See also Quine’s letter from August 14, 1943 to Alonzo Church: “I would hope eventually for
an empirical definition or criterion of synonymy as applied to natural languages.” I thank Thomas
Ricketts for this suggestion.
Carnap’s notes of these discussions are published and examined in Frost-Arnold, Carnap, Tarski,
49
and Quine at Harvard.

Mancosu, “Harvard 1940–1941,” sect. 2.
50
Rudolf Haller, “Alfred Tarski.”

51
“Two Dogmas in Retrospect,” 392.

52
See also Frost-Arnold, “Quine’s Evolution from ‘Carnap’s Disciple’ to the Author of ‘Two Dog-
53
mas,’” 301: “It seems unlikely that Tarski never voiced these views about logic in Quine’s presence
during their year together at Harvard.”
Next to Tarski’s influence, a second series of events seems to have contributed
significantly to Quine’s adoption of a wide-scope holism, viz. his triangular
correspondence with Morton White and Nelson Goodman in 1947. In a series of
letters, Quine, White, and Goodman discussed, among other things, the prospects
of Quine’s search for a behavioristically acceptable definition of analyticity, a
definition Quine still thought was needed in order to account for logical and
mathematical knowledge. In May 1947, White asked Quine to comment on a
manuscript that he would later publish as “On the Church-Frege Solution of
the Paradox of Analysis.” Briefly put, the paradox runs as follows. Consider the
following two statements:
(1) The attribute of being a brother is identical with the attribute of being a
male sibling.
(2) The attribute of being a brother is identical with the attribute of being a
brother.
Intuitively, (1) is informative whereas (2) is not. Yet, if (1) is true, then both
statements say the same thing.
Quine, who had already corresponded with White on the paradox in 1945,
suggests that the paradox might be solved using C. I. Lewis and Carnap’s distinction
between intensional and structural synonymy. The problem with this solution,
however, is that Quine did not know of any behavioristically acceptable definition
of intensional synonymy. Still, Quine’s letter shows that he had not yet given up
hope of finding such a definition:
It’s bad that we have no criterion of intensional synonymy; still, this frankly and
visibly defective basis of discussion offers far more hope of clarity and progress, far
less danger of mediaeval futility, than does the appeal to attributes, propositions, and
meanings. (Goodman, Quine, and White, “Triangular Correspondence,” 339–40)
In response to both White and Quine, however, Goodman defended a much

more stringent position than Quine, arguing that “the lack of any behavioristic
criterion (or even the dimmest suggestion as to how one might be set up) is a sign
that we are not at all clear as to what it is that we have to define.” According to
Goodman, the whole project of seeking acceptable definitions of analyticity and
synonymy was to be rejected: “when Van uses a term and hopes for a behavioristic
criterion he can’t vaguely outline, he is employing a meaningless mark or noise on
the ground that he needs it . . . and hopes that a meaning will be found for it.”54
Quine, obviously not very happy to be placed in the intensionalist camp by
Goodman (“I have always been all for extension, with the world against me”),
responded by backing Goodman’s position. Quine now granted that he also
“doesn’t know how to apply ‘analytic,’ much less define it.”55 Goodman then, in
his final letter, urges Quine to give up on the project of defining analyticity and
to accept that the analytic-synthetic distinction simply cannot be drawn:
Nelson Goodman, W. V. Quine, and Morton White, “A Triangular Correspondence in 1947,” 343.
54
Goodman, Quine, and White, “A Triangular Correspondence in 1947,” 353–54.

55
If Van agrees that he not only doesn’t know how to define “analytic” but doesn’t know
how to apply it either, what is it that he is hoping to find a behavioristic definition
for? . . . [H]e is looking for a behavioristic definition for which the test of adequacy
will presumably be in accordance with a usage which he doesn’t have before him.
(Goodman, Quine, and White, “A Triangular Correspondence in 1947,” 356–57)
Of course, it is a matter of speculation how instrumental Goodman’s pressure

was.56 Still, the fact is that Quine did adopt a wide-scope holism shortly after
Goodman’s final letter.57
6. unregenerate realism
Let me briefly recap the main conclusions of our discussion thus far. Quine’s
naturalism, as defined by NT and SI, is supported by three mutually reinforcing
commitments: empiricism, holism, and realism. I have shown that Quine defended
a strictly behavioristic version of empiricism and a narrow-scope holism from the
very beginning of his career. These commitments led him to seek a behavioristically
acceptable definition of analyticity, such that we might explain our logical and
mathematical knowledge in an empirically satisfying way. Sometime in the late
1940s, Quine, probably influenced by Tarski and Goodman, found a solution in
extending his holism to logic and mathematics, thereby eliminating the need for
a behavioristic explication of analyticity.
The resulting wide-scope holism Quine defends in “Two Dogmas” not only
constituted a break with Carnapian logical empiricism; it also represents a major
step in Quine’s growing dissatisfaction with first-philosophical perspectives and
hence in the development of his naturalism. For even though an analyticity-based
account of logical and mathematical knowledge is in line with the empiricist thesis
that all our knowledge about the world is ultimately based on sense experience,
such an account still aims to justify this logical and mathematical knowledge from
an extra-scientific perspective.58 Where Quine, in his “Lectures on Carnap,” was
still defending the claim that an analyticity-based approach “has the importance of
enabling us to pursue foundations of mathematics and the logic of science without
Especially Creath, Dear Carnap, Dear Van, 35, and Isaac, “Missing Links,” 275, emphasize the
56
importance of the triangular correspondence to Quine’s rejection of the analytic-synthetic distinction. 

In “Animadversions on the Notion of Meaning,” given as a talk at a colloquium at the University
57
of Pennsylvania in 1949 (where Goodman was based at the time), Quine presents many of his argu-
ments from “Two Dogmas” against standard definitions of analyticity and proposes something very
close to his now famous alternative:  
Another view, not distinguishing the [linguistic and factual] components: we have our
sense experience, and our own system of beliefs. . . . But it is underdetermined by experi-
ence. System as a whole must conform to experience along periphery; but disconformities
can be repaired each by any of many changes of the system. We choose by two canons: 1)
maximum elegance of whole system, 2) maximum conservationism. By 2), the more central
principles resist change the more. These might be called the more analytic: matter of degree.
(“Animadversions on the Notion of Meaning,” 155)
See also Gary Ebbs, “Carnap and Quine on Truth by Convention,” 218n27: the problem with an
58
analyticity-based account of logical and mathematical knowledge is that a definition of analyticity in a

certain language system “is designed to explicate a conception of justification for accepting statements
that is independent of the statement’s explanatory contribution to a scientific theory—a conception
of justification that [the naturalistic] Quine associates with first philosophy.”
encountering extra-logical questions as to the source of the validity of our a priori
judgments,”59 the later Quine would reject any attempt to provide our logical and
mathematical knowledge with an extra-scientific foundation.
Still, as I have argued in section 2, the holistic empiricism defended in
“Two Dogmas of Empiricism” does not yet constitute a completely naturalistic
perspective; without his unregenerate realism, Quine’s position was still compatible
with the instrumentalist pragmatism of James, Schiller, and Dewey. Indeed,
Quine’s remarks in “Two Dogmas of Empiricism” about physical objects being
“comparable, epistemologically, to the gods of Homer”60 are often interpreted
as deeply instrumentalist. The problem, as we have seen in section 4, is that
Quine had not yet found a way to naturalize epistemology. Even though Quine
recognized that we are committed to both physical and abstract objects via his
criterion of ontological commitment, he had not yet found a way to get rid of the
transcendental, distinctively epistemological point of view relative to which our
physical and mathematical objects are myths.
The final stage in the development of Quine’s naturalism, therefore, was to find
a consistent way to reject the idea that we can picture the ‘epistemological point of
view’ as a transcendental perspective which potentially undermines realism about
physical and mathematical objects. In the final two sections, I show that we can
distinguish two phases in this process. First, in the late 1940s, Quine abandoned
his attempts to develop a nominalistically acceptable account of mathematics,
accepting that our commitments to physical and mathematical objects are
epistemically on a par. Second, around 1952, Quine started defending the view that
even our phenomenalistic posits, presupposed in most traditional epistemological
perspectives, are not in any sense more basic than our mathematical and physical
posits, thereby removing the final reason not to adopt a full-blooded naturalism.
7. from nominalism to realism

Let me start by considering Quine’s evolving views on the ontological status of
abstract objects. In the 1930s and 1940s, Quine was actively seeking a nominalistically
acceptable account of mathematics. In his intellectual autobiography, Quine
confirms that he already “felt a nominalist’s discontent with classes” when he was
visiting Vienna and Prague in the early 1930s.61 This uneasiness with abstract
objects resulted in a series of lectures and papers which ultimately culminated
in “Steps toward a Constructive Nominalism,” a paper he wrote with Goodman.62
For our present purposes, it is particularly interesting to examine Quine’s reasons
for seeking a nominalist interpretation of mathematics. Reviewing his lectures
and papers, it becomes clear that these reasons were at least partly philosophical.
In his 1946 lecture on nominalism, for example, Quine distinguishes between a
‘mental’ and a ‘physical’ version of nominalism, the former allowing only mental
Quine, “Lectures on Carnap,” 66.

59
“Two Dogmas of Empiricism,” 44.

60
“Autobiography of W. V. Quine,” 14.

61
For a detailed historical account of Quine’s ideas about nominalism, see Lieven Decock, Trad-
62
ing Ontology for Ideology, ch. 2, Mancosu, “Quine and Tarski on Nominalism,” and Parsons, “Quine’s
Nominalism.”
and the latter allowing only physical particulars, and sketches the motives behind
these views:
In the mental case [the nominalist’s] motive may be an extreme sensationalism: what
we are presented with are sensory events, and it is unphilosophical to assume entities
beyond them, in particular universals. In the physical case, his mentality is likely to be
that of Lord Kelvin, who insisted that he did not understand a process until it was
reduced to terms of impact of bodies like billiard balls. . . . Modern physics may seem
to have cut the ground from under this physical type of nominalist, in abandoning
even Kelvin’s billiard balls . . . [b]ut the nominalist is capable of surviving this. . . .
[T]he nominalist reserves the right to refurbish this conceptual scheme . . . and to
produce a substitute conceptual scheme which, while still theoretically adequate to the
physicist’s purposes, will not countenance any entities beyond those whose existence
it is within the physicist’s professional competence to assert. (“Nominalism,” 17–18)
Clearly, Quine here has not yet fully rejected first philosophy. Even though our
best scientific theories quantify over abstract objects, there are philosophical
reasons for either dismissing entities beyond our primary sense experiences or for
refurbishing the physicist’s conceptual scheme in nominalistically acceptable terms.
This first-philosophical attitude is expressed even clearer a few paragraphs earlier,
where Quine responds to the objection that classical mathematics indispensably
quantifies over abstract objects:
Now surely classical mathematics is part of science; and I have said that universals have
to be admitted as values of its variables; so it follows that the thesis of nominalism is
false. What has the nominalist to say to this? He need not give up yet; not if he loves his
nominalism more than his mathematics. He can make his adjustment by repudiating as
philosophically unsound those parts of science which resist his tenets; and his position
remains strong so long as he can persuade us that these rejected parts of science are
neither intrinsically desirable as ends nor necessary as means to other parts which
are intrinsically desirable. (“Nominalism,” 17, my emphasis)
When Quine later specifies that the “intrinsically desirable end” of science is
effective prediction, it becomes clear that his position here is still compatible with
the first-philosophical instrumentalist’s view that theoretical posits beyond those
needed for effective prediction are merely useful fictions.63
Even though Quine actively sought a nominalist interpretation of classical
mathematics, he was never satisfied with the results of his endeavors.64 No doubt,
this growing pessimism about the possibility of fulfilling the project played an
important role in Quine’s rejection of nominalism in the years after the publication
of “Steps toward a Constructive Nominalism.”65 A 1948 letter to J. H. Woodger,
See also John P. Burgess, “Cats, Dogs, and so on,” 63: “the 1946 lecture gives no real reason
63
that I can see why a nominalist should not be satisfied with instrumentalism.” Burgess also notes that
Quine’s arguments in the lecture are still first-philosophical: “we find Quine light-years away from the
principle professed later . . . that epistemology should be ‘naturalized,’ with the philosopher becom-
ing a citizen of the scientific community. Quine’s epistemology at this stage is thoroughly ‘alienated,’
with the philosopher remaining a foreigner, passing judgment from the outside on [the] soundness
of its work” (Burgess, “Cats, Dogs, and so on,” 61).
In his autobiography, Quine explains how he and Goodman failed to give a complete nominalist
64
account of proof theory, which assumes “strings of [s]igns without limit of length, whereas our program
could countenance them only insofar as physically realized” (The Time of My Life, 198).
When exactly Quine completely dispensed with nominalism turns out to be quite difficult to
65
determine. See Decock, Trading Ontology for Ideology, sect. 2.3.

however, indicates that holism also played an important role in Quine’s rejection
of nominalism:
I suppose the question what ontology to accept is in principle similar to the question
what system of physics or biology to accept: it turns finally on the relative elegance
and simplicity with which the theory serves to group and correlate our sense data. . . .
Now the positing of abstract entities (as values of variables) is the same kind of thing.
As an adjunct to natural science, classical mathematics is probably unnecessary; still it
is simpler and more convenient than any fragmentary substitute that could be given
meaning in nominalistic terms. Hence the motive—and a good one—for positing
abstract entities (which classical mathematics) needs. . . . These very relativistic and
tolerant remarks differ in tone from passages in my paper with Goodman and even
in my last letter, I expect. My ontological attitude seems to be evolving rather rapidly
at the moment.66
Quine’s reflections here indicate that his acceptance of a wide-scope holism

provided him with an argument for allowing abstract objects. If we evaluate our
logical and mathematical theories solely in terms of their contribution to our best
scientific theories, dismissing any extra-scientific justification in terms of analyticity,
then there is no reason not to treat physical and mathematical objects on a par.
After all, both play a similar role in “grouping and correlating our sense data.”
While the early Quine was a realist about physical objects but did not yet want
to fully commit himself to the abstract objects of mathematics for philosophical
reasons, his acceptance of a wide-scope holism in the late 1940s seems to have
removed his reasons not to extend his realism to abstract entities. Indeed, from
his 1948 “On What There Is” onwards, Quine treats physical and mathematical
objects as epistemically on a par.67
8. from phenomenalism to realism

Although Quine, by the late 1940s, adopted a realist position about both physical
and mathematical objects, his realism was still in some sense qualified. While he
endorsed the view that “our acceptance of an ontology is . . . similar in principle
to our acceptance of a scientific theory,” he still had not been able to decisively
dismiss the distinctively epistemological point of view from which “the ontologies
of physical objects and mathematical objects are myths.”68 If we work from within
a phenomenalistic conceptual scheme and only allow sense data, Quine argued,
then we cannot maintain our realism about physical and abstract entities. Quine,
in other words, had not yet incorporated epistemology into his wide-scope
holism, such that his realism about physical and mathematical objects would
become truly “unregenerate.” What he needed was an alternative epistemology
in which the justification of science does not ultimately depend on its relation to
a phenomenalistic conceptual scheme.
The first glimmerings of such an alternative appear in Quine’s “Lectures
on David Hume’s Philosophy.”69 In these lectures, Quine presents Hume as
Quine and Joseph Henry Woodger, “The Quine-Woodger Correspondence.” See also Mancosu,
66
“Quine and Tarski on Nominalism,” 43.

See Quine, From a Logical Point of View, 173–74, and The Time of My Life, 198.
67
“On What There Is,” 16–19.

68
See also Michael Pakaluk, “Quine’s 1946 Lectures on Hume.”

69
claiming that in epistemology it is “[v]ain to seek a rational foundation” and that
“[u]ltimately we can only describe psychological behavior, not justify it.”70 Hume’s
philosophy inspires Quine to distinguish between two epistemological projects,
one that reflects the traditional phenomenalist view that we ought to justify science
in terms of sense data, and one that comes remarkably close to the naturalistic
position he was to adopt in the 1950s (here still called ‘pragmatism’):
Constructive empiricism: explain all meaningful scientific discourse by contextual
definition on the basis finally of reference to direct experience.
Pragmatism: abandon such a project as impossible, and say that our discourse is
merely variously conditioned by experience without being reducible to empirical
terms. Abandon, therefore, empirical criticism of concepts; instead, judge any form
of discourse in terms of its utility—this utility being measured within empirical science
by ordinary empirical methods. (“Lectures on David Hume’s Philosophy,” 135.)
When Quine later came to endorse the view that his holism blocks constructive
empiricism because it entails that a typical theoretical sentence has no distinctive
empirical content of its own, it might have led him to reconsider the pragmatist
option. Indeed, there is some evidence for this. For just before he published “Two
Dogmas of Empiricism,” Quine seems to adopt something like this pragmatist
option in 1950 in “Identity, Ostension, and Hypostasis.” Perhaps it is not a
coincidence that this is also the paper in which he for the first time publicly uses
Neurath’s boat metaphor, the analogy he already used in his early notes and would
later often use to illustrate his naturalism:
we must not leap to the fatalistic conclusion that we are stuck with the conceptual
scheme that we grew up in. We can change it bit by bit, plank by plank, though
meanwhile there is nothing to carry us along but the evolving conceptual scheme
itself. The philosopher’s task was well compared by Neurath to that of a mariner
who must rebuild his ship on the open sea. . . . Our standard for appraising basic
changes of conceptual scheme must be, not a realistic standard of correspondence
to reality, but a pragmatic standard.71 (“Identity, Ostension, and Hypostasis,” 78–79)
By 1950, therefore, Quine had adopted something like Hume’s pragmatic

naturalist epistemology; he endorsed the view that we cannot validate science by
translating its concepts into the epistemologically more basic terms of sense data.
Still, even in “Two Dogmas” Quine talked about “sense data” in describing the
evidential boundaries of his newly developed “empiricism without the dogmas,”72
suggesting that he still had not conclusively dispensed with the instrumentalism
that is compatible with such a phenomenalistic epistemology.73 The very last stage
“Lectures on David Hume’s Philosophy,” 134.
70
The fact that Quine often uses Neurath’s metaphor to illustrate his naturalism might suggest that
71
Neurath’s writings played an important role in Quine’s development. Quine himself, however, claims
there was no such influence. In a letter from April 18, 1986 to Dirk Koppelberg on the latter’s 1990 book
on the close relation between his and Neurath’s philosophy, for example, Quine writes: “my reading of my
predecessors has been very sporadic and inadequate. I was aware superficially of my affinity with Neurath,
as you know, and I am glad now to see the degree to it and the detail. I was not appreciably influenced by
him at the time; I had to grow into the point of view on my own, away from Carnap” (my transcription).
“Two Dogmas of Empiricism,” 44.
72
See also, Murphey, The Development of Quine’s Philosophy, 88–89, 92–93. Murphey especially points
73
to Quine’s adoption of a physicalistic definition of observation sentences as his motivation for dispensing
with phenomenalism. Quine first introduces the notion of ‘observation sentence’ in Word and Object.
in the development of Quine’s naturalism, therefore, consists in his adoption of
the view that sense data are not epistemologically prior after all. This breakthrough
finds its origin in “On Mental Entities,” a paper Quine presented in 1952. In this
paper, Quine sets out to answer the question of “whether we should affirm or deny
that there are such things as . . . immediate, subjective experiences.”74 Against the
phenomenalist picture, Quine here argues that such sense data are posits too:
the notion of pure sense datum is a pretty tenuous abstraction, a good deal more
conjectural than the notion of an external object. . . . Epistemologists have wanted to
posit a realm of sense data, situated somehow just me-ward of the physical stimulus, for
fear of circularity: to view the physical stimulation rather than the sense datum as the
end point of scientific evidence would be to make physical science rest for its evidence
on physical science. But if with Neurath we accept this circularity, simply recognizing
that the science of science is a science, then we dispose of the epistemological motive
for assuming a realm of sense data.75 (“On Mental Entities,” 225–26)
Quine here for the first time adopts a thoroughly naturalistic point of view.
Epistemology is not a distinctively philosophical project responsible for validating
our scientific theories and for blocking an unregenerate realism about physical
and mathematical objects. Rather, Quine now endorses the view that epistemology
is itself a science, and therefore cannot be conducted from some transcendental
science-independent perspective. Quine, in other words, had adopted the view
that the idea that “any evidence for science has its end points in the senses . . . is
an insight which comes after physics, physiology, and psychology, not before.”76
As I have argued in section 2, the crucial argument underlying Quine’s realism
is the idea we cannot ask about the ‘reality’ and ‘truth’ of our scientific posits and
theories in a distinctively philosophical way without stripping those concepts of
their intelligibility. Our notions of ‘reality’ and ‘truth’ themselves depend on our
scientific conceptual scheme. Not surprisingly, this argument also first appears in
“On Mental Entities.” Phenomenalist epistemologists, Quine argues, regard the
realm of sense data as somehow more “real” than the external objects that are
posited in order to organize our experiences. Quine now believes, however, that
“it is a mistake to seek an immediately evident reality, somehow more immediately
evident than the realm of external objects”:
Everything, of course, is real; but there are sheep and there are no unicorns . . . there
are odd numbers and there are no even primes other than 2. Such is the ordinary

74
See also “The Place of a Theory of Evidence,” a lecture from October 7, 1952: “We would do
75
well to recognize that in seeking to isolate sense data we are not plumbing the depths of reality; we are
engaged rather in empirical psychology, associating physical stimuli with human responses. From the
laws of this science the sense data as intermediary hypothetical entities may, however, be deleted; they
will not be missed.” The seeds of Quine’s rejection of sense data as epistemologically neutral build-
ing blocks can already be found in his graduate paper “Concepts and Working Hypotheses” (Quine
1931): “My immediate experience, rather than consisting of raw material to be interpreted, is already
seething with interpretation . . . analysis of a given experience can yield any other experience which
is, in any full sense, the ‘bare datum’ of the former experience; any such analysis is, rather, merely
a further interpretation.” I thank an anonymous referee for drawing my attention to this passage.
76
usage of the word ‘real.’ . . . Failing some aberrant definition which is certainly not
before us, this is the only usage we have to go on.77 (“On Mental Entities,” 225)
Quine, in other words, fully embraced a naturalistic point of view; he adopted a

perspective immanent to our scientific conceptual scheme and principally rejected
any transcendental perspective on reality. His wide-scope holism now not only
included logic and mathematics, but also our ideas about the sensory basis of
science. Combined with his argument for an unregenerate realism, Quine had
definitively boarded Neurath’s boat.
9. conclusion
In this paper, I have given a preliminary account of how Quine arrived at his
naturalism. I have argued that Quinean naturalism is supported by three substantive
commitments—empiricism, holism, and realism—and have pieced together the
evolution of his position by examining the origins of these commitments in his
work. Building on Quine’s unpublished notes and early writings as well as on
the existing literature on his early views, I have argued that, although Quine had
from the early beginnings defended a behavioristic version of empiricism and a
narrow-scope holism, it was not until the late 1940s that he, probably influenced
by Tarski and Goodman, realized that he could broaden his holism to include
logic and mathematics.
Once Quine had adopted this wide-scope holism, he gradually started grasping
the radically naturalistic consequences of his position. First, Quine came to reject
attempts to find an analyticity-based account of our logical and mathematical
knowledge, defending the view that there is no need to justify this knowledge
outside its contribution to our overall scientific theories. Secondly, Quine
abandoned his attempts to seek a nominalistically acceptable interpretation of
mathematics, a project that was at least partially guided by first-philosophical
motives. Instead, he came to defend the view that on a rigorously holistic picture
of inquiry there is no reason not to treat physical and mathematical objects as
epistemically on a par. Thirdly, from 1952 onwards, Quine adopted the view that
a similar move could be made with respect to epistemology; there is no distinct
epistemological point of view which deals with objects (sense data) that are in any
sense more ‘real’ than the objects posited in the sciences. Rather, Quine adopted
the view that sense data are theoretical posits, that there is no meaningful extra-
scientific notion of ‘reality,’ and that the very empiricism he had defended from
the beginning of his career, itself could only be plausibly defended from within
the framework of science.78
See also Quine, “The Scope and Language of Science,” 233: “the terms ‘reality’ and ‘evidence’
77
owe their intelligibility to their applications in archaic common sense.”

I would like to thank Lieven Decock, Gary Ebbs, Hans-Johann Glock, Jeanne Peijnenburg, and
78
Allard Tamminga for their valuable comments on earlier drafts of this paper. Sections of this paper
have been presented at the workshop “Wittgenstein and Quine on Truth, Evidence, and Semantic
Content” at the University of Zurich in September 2014. I would like to thank the audience, especially
Peter Hylton and Thomas Ricketts, for their valuable suggestions. Finally, I would like to thank the
three anonymous referees for the Journal of the History of Philosophy for their useful comments. This
research is funded by the Netherlands Organization for Scientific Research (NWO), Grant 322-20-001.
bibliography and abbreviations
Quine’s Works
Quine’s works are listed chronologically to emphasize the development of his thought. The unpublished
papers, letters, lectures, and notebooks listed below are stored at the Harvard Depository and can be
accessed at Houghton Library. A catalogue of Quine’s unpublished work can be found at http://oasis.
lib. harvard.edu/oasis/deliver/~hou01800.
Quine, W. V. “Miscellaneous Papers.” Unpublished Student Papers, 1925–1931. W. V. Quine Papers

(MS Am 2587): Box 110, Item 3225. Houghton Library, Harvard University.
———. “On the Organization of Knowledge.” Unpublished Student Paper, March 10, 1927. W. V.
Quine Papers (MS Am 2587): Box 110, Item 3225. Houghton Library, Harvard University.
———. “Papers in Philosophy.” Unpublished Student Papers, 1930–31. W. V. Quine Papers (MS Am
2587): Box 111, Item 3236. Houghton Library, Harvard University.
———. “Concepts and Working Hypotheses.” Unpublished Student Paper, March 10, 1931. W. V.
Quine Papers (MS Am 2587): Box 111, Item 3236. Houghton Library, Harvard University.
———. “Lectures on Carnap,” 1934. In Creath, Dear Carnap, Dear Van, 45–103.
———. “Truth by Convention,” 1936. In Quine, The Ways of Paradox, 77–106.
———. “Early Jottings on Philosophy of Language.” Unpublished Notes, 1937–44. W. V. Quine Papers
———. “A Tentative Ontology.” Unpublished Autograph Notes, November 28, 1941. W. V. Quine
Papers (MS Am 2587): Box 108, Item 3169. Houghton Library, Harvard University.
———. “Whitehead and the Rise of Modern Logic,” 1941. In Quine, Selected Logic Papers, 3–36.
———. “Foundations of a Linguistic Theory of Meaning.” Unpublished Manuscript, August 1943.
W. V. Quine Papers (MS Am 2587): Box 108, Item 3169. Houghton Library, Harvard University.
———. “Things.” Unpublished Autograph Note, January 30, 1943. W. V. Quine Papers (MS Am 2587):
Box 108, Item 3169. Houghton Library, Harvard University.
———. “Ontology and Epistemology.” Unpublished Autograph Note, March 27, 1944. W. V. Quine
———. “Ontology Notes.” Unpublished Autograph Notes, November 5, 1944. W. V. Quine Papers
———. “Sign and Object; or, The Semantics of Being.” Unpublished Autograph Note, October 4, 1944.
W. V. Quine Papers (MS Am 2587): Box 108, Item 3169. Houghton Library, Harvard University.
———. “What it Means to Be.” Unpublished Autograph Note, March 19, 1944. W. V. Quine Papers
———. “Ontology, Metaphysics, etc.” Unpublished Notes, 1944–1951. W. V. Quine Papers (MS Am
2587): Box 108, Item 3181. Houghton Library, Harvard University.
———. “Lectures on David Hume’s Philosophy,” 1946. Edited by J. Buickerood. In Quine, Confessions,
36–136.
———. “Nominalism,” 1946. Transcribed by P. Mancosu. In Quine, Confessions, 9–23.
———. “On What There Is,” 1948. In Quine, From a Logical Point of View, 1–19.
———. “Animadversions on the Notion of Meaning,” 1949. In Quine, Confessions, 152–56.
———. “Identity, Ostension, and Hypostasis,” 1950. In Quine, From a Logical Point of View, 65–79.
———. “Two Dogmas of Empiricism,” 1951. In Quine, From a Logical Point of View, 20–46.
———. “On Mental Entities,” 1952. In Quine, The Ways of Paradox, 221–27.
———. “The Place of a Theory of Evidence.” Unpublished Lecture, 1952. W. V. Quine Papers (MS
Am 2587): Box 104, Item 3011. Houghton Library, Harvard University.
———. “Mr. Strawson on Logical Theory,” 1953. In Quine, The Ways of Paradox, 137–57.
———. “The Scope and Language of Science,” 1954. In Quine, The Ways of Paradox, 228–45.
———. “Posits and Reality,” 1955. In Quine, The Ways of Paradox, 246–54.
———. Word and Object. Cambridge: The MIT Press, 1960.
———. From a Logical Point of View. 2nd Edition. London: Oxford University Press, 1961.
———. “Stimulus and Meaning,” 1965. In The Isenberg Memorial Lecture Series, edited by R. Suter, 39–61.
East Lansing: The Michigan State University Press, 1969.
———. “Ontological Relativity.” Unpublished Lecture. Prepared March 1967. W. V. Quine Papers (MS
Am 2587): Box 103, Item 2994. Houghton Library, Harvard University.
———. “Ontological Relativity,” 1968. In Quine, Ontological Relativity, 26–68.
———. “Epistemology Naturalized,” 1969. In Quine, Ontological Relativity, 69–90.
———. Ontological Relativity and Other Essays. Cambridge, MA: Harvard University Press, 1969.
———. “Five Milestones of Empiricism,” 1975. In Quine, Theories and Things, 67–72.
———. “On Empirically Equivalent Systems of the World,” 1975. In Quine, Confessions, 228–43.
———. The Ways of Paradox, rev. ed. Cambridge: Harvard University Press, 1976.
———. “Things and Their Place in Theories,” 1981. In Quine, Theories and Things, 1–23.
———. “The Pragmatists’ Place in Empiricism,” 1981. In Pragmatism: Its Sources and Prospects, edited
by R. J. Mulvaney and P. M. Zeltner, 21–39. Columbia: University of South Carolina Press, 1981.
———. Theories and Things. Cambridge: Harvard University Press, 1981.
———. The Time of My Life: An Autobiography. Cambridge: The MIT Press, 1985.
———. “Autobiography of W. V. Quine.” In The Philosophy of W. V. Quine, edited by L. Hahn and P.
Schilpp, 1–46. La Salle: Open Court, 1986.
———. “The Sensory Support of Science,” 1986. In Quine, Confessions, 327–37.
———. “Two Dogmas in Retrospect,” 1991. In Quine, Confessions, 390–400.
———. Pursuit of Truth, rev. ed. Cambridge: Harvard University Press, 1992.
———. “Structure and Nature,” 1992. In Quine, Confessions, 401–6.
———. “Replies to Professor Riska’s Eight Questions,” 1992. In Quine, Quine in Dialogue, 213–15.
———. “Twentieth-Century Logic,” 1994. Interview by G. Borradori. In Quine, Quine in Dialogue, 57–68.
———. “Responses to Articles by Abel, Bergström, Davidson, Dreben, Gibson, and Prawitz,” 1994. In
Quine, Quine in Dialogue, 223–34. [“Responses”]
———. Selected Logic Papers, rev. ed. Cambridge: Harvard University Press, 1995.
———. “Naturalism; Or, Living within One’s Means,” 1995. In Quine, Confessions, 461–72.
———. Confessions of a Confirmed Extensionalist and Other Essays. Edited by Dagfinn Føllesdal and Douglas
Quine. Cambridge: Harvard University Press, 2008. [Confessions]
———. Quine in Dialogue. Edited by Dagfinn Føllesdal, and Douglas Quine. Cambridge: Harvard
University Press, 2008.
———, and Alonzo Church. “The Quine-Church Correspondence,” 1935–1994. W. V. Quine Papers
———, and Dirk Koppelberg. “The Quine-Koppelberg Correspondence,” 1981–1994. W. V. Quine
———, and Joseph Henry Woodger. “The Quine-Woodger Correspondence,” 1938–1982. W. V. Quine
Secondary Sources
Almeder, Robert F. Harmless Naturalism: The Limits of Science and the Nature of Philosophy. Chicago: Open
Court, 1998.
Barrett, Robert B. and Roger F. Gibson, eds. Perspectives on Quine. Oxford: Basil Blackwell, 1990.
Burgess, John P. “Cats, Dogs, and so on.” In Zimmerman, Oxford Studies in Metaphysics, 56–78.
Carnap, Rudolf. Introduction to Semantics. Cambridge: Harvard University Press, 1942.
———. The Logical Syntax of Language. Translated by Amethe Smeaton. London: Routledge & Kegan
Paul, 1937.
Creath, Richard, ed. Dear Carnap, Dear Van. Berkeley: University of California Press, 1990.
———. “The Initial Reception of Carnap’s Doctrine of Analyticity.” Noûs 21 (1987): 477–99.
Decock, Lieven. Trading Ontology for Ideology: The Interplay of Logic, Set Theory and Semantics in Quine’s
Philosophy. Dordrecht: Kluwer Academic Publishers, 2002.
Dreben, Burton. “Quine.” In Barrett and Gibson, Perspectives on Quine, 81–95.
Ebbs, Gary. “Carnap and Quine on Truth by Convention.” Mind 120 (2011): 193–237.
Føllesdal, Dagfinn. “Developments in Quine’s Behaviorism.” American Philosophical Quarterly 48 (2011):
273–82.
———, ed. Philosophy of Quine, Volume 2: Naturalism and Ethics. New York: Garland Publishing, 2001.
Frost-Arnold, Greg, ed. Carnap, Tarski, and Quine at Harvard: Conversations on Logic, Mathematics, and
Science. Chicago: Open Court, 2013.
———. “Quine’s Evolution from ‘Carnap’s Disciple’ to the Author of ‘Two Dogmas.’” HOPOS: The
Journal of the International Society for the History of Philosophy of Science 1 (2011): 291–316.
Gibson, Roger F. Jr. Enlightened Empiricism: An Examination of Quine’s Theory of Knowledge. Tampa:
University of South Florida Press, 1988.
Godfrey-Smith, Peter. “Quine and Pragmatism.” In Harman and Lepore, A Companion to W. V. O.
Quine, 54–68.
Goodman, Nelson, and W. V. Quine. “Steps towards a Constructive Nominalism.” Journal of Symbolic
Logic 12 (1947): 105–22.
———, W. V. Quine, and Morton White. “A Triangular Correspondence in 1947.” In A Philosopher’s
Story, edited by M. White, 337–57. University Park: The Pennsylvania State University Press, 1999.
Haller, Rudolf. “Alfred Tarski: Drei Briefe an Otto Neurath.” Grazer Philosophische Studien 43 (1992):
1–32.
Harman, Gilbert and Ernest Lepore, eds. A Companion to W. V. O. Quine. Chichester: Blackwell, 2014.
Hylton, Peter. “‘The Defensible Province of Philosophy’: Quine’s 1934 Lectures on Carnap.” In Future
Pasts: The Analytic Tradition in Twentieth-Century Philosophy, edited by J. Floyd and S. Shieh, 57–76.
New York: Oxford University Press, 2001.
———. Quine. New York: Routledge, 2007.
———. “Quine’s Naturalism Revisited.” In Harman and Lepore, A Companion to W. V. O. Quine, 148–62.
Isaac, Joel. “Missing Links: W. V. Quine, the Making of ‘Two Dogmas,’ and the Analytic Roots of Post-
Analytic Philosophy.” History of European Ideas 37 (2011): 267–79.
———. “W. V. Quine and the Origins of Analytic Philosophy in the United States.” Modern Intellectual
History 2 (2005): 205–34.
Katz, Jerrold J. “The Refutation of Indeterminacy.” In Barrett and Gibson, Perspectives on Quine, 177–97.
Koppelberg, Dirk. Die Aufhebung der analytischen Philosophie: Quine als Synthese von Carnap und Neurath.
Frankfurt: Suhrkamp Verlag, 1990.
Koskinen, Heikki J. and Sami Pihlström. “Quine and Pragmatism.” Transactions of the Charles S. Peirce
Society, 42 (2006): 309–46.
Lewis, Clarence Irving. Mind and the World-Order. New York: Charles Scribner’s Sons, 1929.
Macarthur, David. “Quinean Naturalism in Question.” Philo 11 (2008): 1–14.
Mancosu, Paolo. “Harvard 1940–1941: Tarski, Carnap and Quine on a Finitistic Language of
Mathematics for Science.” History and Philosophy of Logic 26 (2005): 327–57.
———. “Quine and Tarski on Nominalism.” In Zimmerman, Oxford Studies in Metaphysics, 22–55.
Morris, Sean. “Quine, Russell, and Naturalism: From a Logical Point of View.” Journal of the History of
Philosophy 53 (2015): 133–55.
Murphey, Murray G. The Development of Quine’s Philosophy. Dordrecht: Springer, 2012.
Pakaluk, Michael. “Quine’s 1946 Lectures on Hume.” Journal of the History of Philosophy 27 (1989):
445–59.
Parsons, Charles. “Quine’s Nominalism.” American Philosophical Quarterly 48 (2011): 213–28.
Sinclair, Robert. “Quine and Conceptual Pragmatism.” Transactions of the Charles S. Peirce Society 48
(2012): 335–55.
Tennant, Neil. “Carnap and Quine.” In Logic, Language, and the Structure of Scientific Theories, edited
by Wesley Salmon and Gereon Wolters, 305–44. Pittsburgh: University of Pittsburgh Press, 1994.
Verhaegh, Sander. “Blurring Boundaries: Carnap, Quine, and the Internal-External Distinction.”
Erkenntnis (forthcoming).
———. “Quine’s Argument from Despair.” British Journal for the History of Philosophy 22 (2014): 150–73.
———. Rafts, Boats, and Cruise Ships: Naturalism and Holism in Quine’s Philosophy. PhD Diss., University
of Groningen, 2015.
Weir, Alan. “Quine’s Naturalism.” In Harman and Lepore, A Companion to W. V. O. Quine, 114–47.
White, Morton. “On the Church-Frege Solution of the Paradox of Analysis.” In White, From a Philosophical
Point of View, 112–15. Princeton: Princeton University Press, 2005.
———, and Alfred Tarski. “A Philosophical Letter of Alfred Tarski.” The Journal of Philosophy 84 (1987):
28–32.
Zimmerman, Dean W., ed. “Quine’s 1946 Lecture on Nominalism: a Symposium.” Oxford Studies in
Metaphysics, Volume 4. Oxford: Oxford University Press, 2008.

Early Development of Quine Naturalism

Uploaded by

Copyright:

Available Formats

Early Development of Quine Naturalism

Uploaded by

Document Information

Original Description:

Original Title

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

Early Development of Quine Naturalism

Uploaded by

Copyright:

Available Formats

Boarding Neurath's Boat: The Early Development of Quine's

Published by Johns Hopkins University Press

For additional information about this article

abstract W. V. Quine is arguably the intellectual father of contemporary natural-

keywords Quine, naturalism, empiricism, holism, realism, archive studies

Quine, “Things and Their Place in Theories,” 21.

Journal of the History of Philosophy, vol. 55, no. 2 (2017) 317–342

(1930–1931), “Early Jottings on Philosophy of Language” (1937–1944), and “Ontology, Metaphysics,

2. quinean naturalism decomposed

Applied to epistemology and metaphysics, NT and SI deliver some of Quine’s most

Pursuit of Truth, 19.

“Things and Their Place in Theories,” 21.

“Structure and Nature,” 405.

Quine, “On Empirically Equivalent Systems of the World,” 230.

Verhaegh, “Quine’s Argument from Despair.”

“On Mental Entities,” 252.

Why does Quine cite realism as a source of naturalism? It is my contention that

Realism, in other words, is an important feature of Quine’s naturalism; without it,

“The Pragmatists’ Place in Empiricism,” 33.

“The Pragmatists’ Place in Empiricism,” 33.

4. quine’s naturalism in the early 1940s

“The Defensible Province of Philosophy.” I follow Creath in describing Quine’s characterization as

In these fragments, Quine is worried about the objections of phenomenalist

5. narrow and wide holism

Indeed, Carnap himself also combined his analyticity-based explanation of mathematical

He says after some reading of my “Syntax” MS:

Mancosu, “Harvard 1940–1941,” 331.

and Quine at Harvard.

Rudolf Haller, “Alfred Tarski.”

“Two Dogmas in Retrospect,” 392.

In response to both White and Quine, however, Goodman defended a much

Goodman, Quine, and White, “A Triangular Correspondence in 1947,” 353–54.

Of course, it is a matter of speculation how instrumental Goodman’s pressure

importance of the triangular correspondence to Quine’s rejection of the analytic-synthetic distinction.

analyticity-based account of logical and mathematical knowledge is that a definition of analyticity in a

7. from nominalism to realism

Quine, “Lectures on Carnap,” 66.

“Two Dogmas of Empiricism,” 44.

“Autobiography of W. V. Quine,” 14.

determine. See Decock, Trading Ontology for Ideology, sect. 2.3.

Quine’s reflections here indicate that his acceptance of a wide-scope holism

8. from phenomenalism to realism

“Quine and Tarski on Nominalism,” 43.

“On What There Is,” 16–19.

See also Michael Pakaluk, “Quine’s 1946 Lectures on Hume.”

By 1950, therefore, Quine had adopted something like Hume’s pragmatic

“On Mental Entities,” 221.

Quine, in other words, fully embraced a naturalistic point of view; he adopted a

owe their intelligibility to their applications in archaic common sense.”

Quine, W. V. “Miscellaneous Papers.” Unpublished Student Papers, 1925–1931. W. V. Quine Papers

You might also like

He says after some reading of my “Syntax” MS: 

importance of the triangular correspondence to Quine’s rejection of the analytic-synthetic distinction. 