BPH A

BELGRADE PHILOSOPHICAL ANNUAL
/FILOZOFSKI GODIŠNJAK 27/2014

Institute of Philosophy, Faculty of Philosophy, University of Belgrade
Belgrade, Čika Ljubina 18-20
Belgrade
Year XXVII
YU ISSN 0353-3891
UDK–1
Editor
Slobodan Perović (University of Belgrade)
Editorial Board
Miloš Arsenijević (University of Belgrade)
Jovan Babić (University of Belgrade)
Leon Kojen (University of Belgrade)
Živan Lazović (University of Belgrade)
Timothy Williamson (University of Oxford)
Advisory Board
Berit Brogaard (University of Miami)
Paul Boghossian (NY University)
Bob Hale (University of Sheffield)
Aleksandar Jokić (Portland State University)
Jan Narveson (University of Waterloo)
Georg Meggle (University of Leipzig)
J. Angelo Corlett (San Diego State University)
Howard Robinson (Central European University)
David Charles (University of Oxford)
Managing Editor
Željko Mančić (University of Belgrade)
Belgrade Philosophical Annual online at http://www.f.bg.ac.rs/bpa/index.html

Printed by
„Službeni Glasnik,“ Belgrade
This issue is financially supported by
Ministry of Education, Science and Technological Development of the Republic
of Serbia
BELGRADE PHILOSOPHICAL ANNUAL 27/2014
EDITOR’S NOTICE .............................................................................................. 5

ARTICLES
Timothy How did we get here from there?
Williamson The transformation of analytic philosophy ..................... 7
DELIBERATIVE DEMOCRACY
Jürg Steiner The dynamics of deliberation ............................................ 39
Maria Clara
Jaramillo
Simona Mameli
Milorad Stupar Democracy and truth: a cluster of problems –
an attempt to sort them out ............................................... 49
Ivan Mladenović Deliberative democracy and the
epistemology of disagreement ........................................... 61
ARTICLES
Jan Narveson Democracy: is it both possible and desirable? ................ 71
Nicholas Rescher Oversimplification .............................................................. 85
Werner Sauer Metaphysics EZHO: Aristotle’s Unified Treatise περὶ τῆς
οὐσίας καὶ τοῦ ὄντος, de Essentia et de Ente .................. 93
Sanja Srećković Eduard Hanslick’s formalism and his most influential
contemporary critics ........................................................... 113
Ljiljana Radenović Empirical research into the problem of other minds:
how developmental social psychology can strengthen
the Wittgensteinian attitudinal approach ........................ 135
Drago Đurić Al-Kindi’s and W. L. Craig’s cosmological arguments ... 157
Anand Jayprakash On the role of modal intuition in modal logic ............... 167
Vaidya
EDITOR’S NOTICE
Belgrade Philosophical Annual is published by the Institute for Philosophy,

University of Belgrade. The first issue of the journal appeared in 1988. The
journal accepts submissions of papers on a broad domain of philosophical topics
that meet respectful standards of clarity and scholarly research. Until 2013 most
papers published in the journal were written in Serbian. Since 2013 the Editorial
Board decided to accept submissions in English only. The Board also decided to
change the old version of the name of the journal in English from ‘Philosophy
Yearbook’ to its current name. The name of the journal in Serbian remained
unchanged (‘Filozofski Godišnjak’). The printed edition is published once a year,
while online papers are issued in four annual installments.
ARTICLES
Timothy Williamson Original Scientific Paper

Oxford University UDK 1(091)”19/20”
165.6/.8”19”
HOW DID WE GET HERE FROM THERE? THE

TRANSFORMATION OF ANALYTIC PHILOSOPHY1
Opponents of analytic philosophy often associate it with logical positivism.

From a historical point of view, it is clear that one main strand in the development
of the broad tradition known as ‘analytic philosophy’ was indeed the logical
positivism of the Vienna Circle, with its austerely verificationist principle
of significance and its exclusion of metaphysics as cognitively meaningless.
Another main strand in the development of the analytic tradition, ordinary
language philosophy, tended to be almost equally suspicious of the ways in
which metaphysicians made free with ordinary words, far from the everyday
contexts of use on which their meaning was supposed to depend. Despite that
history, however, recent decades have seen the growth and flourishing of boldly
speculative metaphysics within the analytic tradition. Far from being inhibited by
logical positivist or ordinary language scruples, such analytic metaphysics might
be described by those unsympathetic to it as pre-critical, ranging far outside the
domain of our experience, closer in spirit to Leibniz than to Kant.
1 This paper derives from a talk given at the Faculty of Philosophy of Belgrade University
in September 2014. I thank Professor Miroslava Trajkovski for the suggestion that I might
depart from my usual practice and give a talk on an historical theme, based in part on my
own experience. Thanks to the audience for constructive discussion, and to Peter Vallentyne
for spotting some bad typos. Additional thanks go to Professor Slobodan Perović for inviting
me to write up the paper for publication in this journal. He also pointed out the thematic
connection with Peter Strawson’s 1977 lecture at Belgrade University on the nature of analytic
philosophy, subsequently translated into Serbo-Croat (as it then was) and published in
Theoria (Strawson 1977). On that visit to Yugoslavia, Strawson lectured in Belgrade, Zagreb,
and Sarajevo. He later commented: ‘I registered a certain difference in atmosphere in the
three places. At least in academic circles the intellectual style seemed relatively untrammelled
in Belgrade and Zagreb, though the political tone was different. In Sarajevo, where I was
only allowed to give one of my two scheduled lectures and had minimal contact with fellow
academics, one perhaps time-serving young man in my audience suggested that my lecture
revealed an essentially bourgeois outlook. I replied “But I am bourgeois–an elitist liberal
bourgeois”. My interpreter commented, sotto voce, “They envy you”’ (Strawson 1998, p. 14).
The contrast between Strawson’s account of analytic philosophy and the present one, almost
forty years later, may be instructive.
8 BELGRADE PHILOSOPHICAL ANNUAL Vol. XXVII (2014)
How did a species of philosophy with so much anti-metaphysics in its gene

pool evolve so quickly to the opposite extreme? Enough time has passed for us
now to start achieving the historical perspective necessary to answer the question
in a systematic way. That project is too extensive to be properly carried out in less
than a book. In this paper, I attempt no more than to make some informal and
unsystematic remarks on the transformation of analytic philosophy. Especially in
section II, I write as someone who lived through the latter stages of the process,
and concentrate on the parts of which I have the closest knowledge. That will at
least provide some sense of what it was like to experience the transformation at
the time, from a broadly sympathetic point of view. I close with a few sketchy
remarks on the historiography of recent analytic philosophy, in section III.
I
The central, most influential figure in the development of analytic
metaphysics over the final quarter of the twentieth century, and the contemporary
philosopher most cited within recent analytic philosophy, is undoubtedly David
Lewis, also known as ‘the machine in the ghost’ for his eerie computational
power, mechanical diction, faint air of detachment from ordinary life, and
beard from another era (by contrast with ‘the ghost in the machine’, Gilbert
Ryle’s summary description of the immaterial Cartesian ego in the clockwork
Cartesian body). The prize specimen of Lewis’s speculative metaphysics is in turn
his notorious doctrine of modal realism, according to which there are infinitely
many possible worlds, mutually disconnected spatiotemporal systems each as
real and concrete as our own actual world (Lewis 1986). For Lewis, strictly and
literally there are talking donkeys, because there could have been talking donkeys
(as we may all agree), so some possible world contains talking donkeys, which
are just as real, alive, and made of flesh and blood as any donkey you have ever
seen. Of course, those other worlds are not open to our observation; there are
no trans-world telescopes. Lewis postulates them because they follow from his
modal realism, which he regards as the best theory of possibility, necessity, and
related phenomena, in respect of simplicity, strength, elegance, and explanatory
power: to use C.S. Peirce’s term broadly, Lewis’s argument for modal realism is
abductive. In a way, Lewis takes non-actual possible worlds even more seriously
than did Leibniz, for whom they are merely unrealized ideas in the mind of God.
Leibniz’s God realized only the best of all possible worlds, whereas all of Lewis’s
possible worlds are equally realized. We can take Lewis’s modal realism as a case
study for the resurgence of speculative metaphysics in contemporary analytic
philosophy.
Some philosophers treat any appeal to possible worlds at all as a metaphysical
extravagance. But that is a mistake, for some theories treat possible worlds as
merely abstract objects or representations, harmlessly built from this-worldly
materials. Indeed, Rudolf Carnap, the logical positivist anti-metaphysician par
excellence, explicitly compared to Leibniz’s possible worlds his state-descriptions,
maximal consistent classes of sentences used in his semantics for languages with
Timothy Williamson: How did we get here from there? Te transformation of analytic philosophy 9
modal operators such as ‘possibly’ and ‘necessarily’. He also compared them

with the possible states of affairs in Wittgenstein’s Tractatus Logico-Philosophicus
(Carnap 1947, p. 9). For Carnap, necessity is a purely intra-linguistic matter of
truth guaranteed by meaning, and possibility is a correspondingly semantic form
of consistency. Lewis’s modal realism has always been a minority view, even
amongst those analytic metaphysicians who work with possible worlds of some
sort. It is an example of extreme metaphysics.
Where did Lewis’s modal realism come from? It already appears in one of
his earliest published papers (Lewis 1968). Exciting developments in modal logic,
culminating in the work of Saul Kripke, had already made the idea of possible
worlds central to the understanding of languages with modal operators (Kripke
1963). ‘Necessarily’ is understood as ‘in every possible world’ and ‘possibly’ as ‘in
some possible world’. For technical mathematical reasons, it turned out to be more
fruitful not to equate possible worlds with Carnap’s state-descriptions, or other
such representational entities, but instead to leave their nature unconstrained
when characterizing models for the modal language (Williamson 2013, pp.
81–4). That did not enforce a more metaphysically speculative conception of
possible worlds, but it made space for one.
Another significant factor was the development of tense logic, above all by
Arthur Prior. He was acutely aware of the strong structural analogies between
modal logic and tense logic, with possible worlds playing the same role in the
modal case as moments of time play in the temporal case (Prior 1957). An
orthographically identical operator could be read as ‘necessarily’ in modal
logic and as ‘always’ (or ‘always in the past’ or ‘always in the future’) in tense
logic. When one read formulas of the logic in temporal terms, they expressed
blatantly metaphysical principles about the structure of time and of existence in
time. Those readings provided templates for more metaphysical readings of the
same formulas in modal terms, on which they expressed analogous metaphysical
principles about the structure of possibility and of possible or necessary
existence. Carnap’s intra-linguistic modalities could then be replaced by Kripke’s
metaphysical modalities, which concern how things really could have been.
But developments within modal logic alone cannot fully explain Lewis’s
modal realism. For Kripke and other leading modal logicians did not go down
the modal realist road. Indeed, Lewis’s modal realist semantics for modal
languages introduces messy complications that from a purely technical point of
view are quite unmotivated, even by Prior’s analogy between time and modality.
In particular, Lewis postulates that no individual exists in more than one world,
the analogue of the highly implausible postulate that no individual exists for
more than one moment. In order to make sense of the common sense idea
that one could have acted differently, Lewis then has to introduce an elaborate
theory of counterparthood relations between distinct but similar individuals in
different possible worlds. Those complications were motivated by Lewis’s prior
commitments.
At this point, one must note that Lewis’s doctoral supervisor at Harvard
in the mid–1960s was Willard van Orman Quine, the leading critic of modal
logic. Quine was especially critical of quantified modal logic, as developed by

Carnap and Ruth Barcan Marcus, which allows one in effect to reason about the
properties and relations that particular individuals could have, as opposed to just
the general states of affairs that could obtain. Originally, Quine had tried to prove
that quantified modal logic is technically flawed to the point of incoherence, that
it collapses the modal distinctions between possibility, necessity, and actuality.
As it became clear that his purely formal arguments were technically flawed,
Quine gradually switched his line of attack to the informal intelligibility of
quantified modal formulas. His standard of intelligibility in logic was austere:
first-order non-modal logic, roughly, that of the logical constants ‘not’, ‘and’, ‘or’,
‘everything’, ‘something’, and ‘is’. For Quine, logic is first-order non-modal logic.
Lewis assumed modal realism because it permits the reduction by translation
of a quantified modal language to a first-order non-modal language in which
one talks about worlds and individuals in those worlds. Crucially, Lewis’s modal
realism gave him a way of informally explaining what a possible world is in non-
modal terms: roughly, a spatiotemporal system; the individuals in such a system
are spatiotemporally connected to each other and to nothing outside the system.
Lewis thereby aimed to make quantified modal logic intelligible by his teacher’s
standards. Since much ordinary discourse in natural language involves expressive
resources at least as great as those of quantified modal logic – ‘can’ and ‘must’
are common words, to say the least – Lewis’s procedure is also motivated by
a principle of charity (Lewis 1974), which Quine explicitly endorsed: prefer an
interpretation of a natural language on which speakers are being sensible to one
on which they are being silly (Quine 1960, p. 59). Possible worlds other than
our own and their inhabitants are needed to make true ordinary statements
about what could have been but isn’t. Lewis gave an ingenious solution to the
Quinean problem of charitably interpreting quantified modal discourse, on its
own Quinean terms, even though Quine rejected the gift.
Quine’s lack of interest in modal realism was no anti-metaphysical stance. He
did as much as anyone to put ontology as a branch of metaphysics on the map of
analytic philosophy, and his conception of philosophy as continuous with natural
science overtly involved a naturalistic approach to metaphysical theorizing. One
might rather be tempted to suggest that he rejected modal realism because it
lacked support from natural science. However, Quine did not require each point
of a metaphysical theory to receive its own specific support from natural science.
For instance, although he took mathematics – with its ontological commitment
to abstract objects such as sets or numbers – to be holistically justified by its
applications in natural science, he was well aware that the power of the standard
axioms of set theory goes far beyond the needs of natural science, but still
regarded them as a legitimate rounding out of the fragment actually used in
scientific applications, justified by its simplicity, elegance, and other such virtues.
Thus Quine’s justification of mathematics is abductive, in a similar spirit to
Lewis’s justification of modal realism. No doubt, if Quine had felt some tension
between modal realism and current natural scientific theory, he would have
treated that as a good reason to reject modal realism, but so might Lewis himself.
However, neither of them seems to have felt such a tension. In my view, there is
in fact such a tension, or inconsistency, but the argument for it must be made
with some delicacy and was not generally recognized at the time (Williamson
2013, pp. xii, 17 and 2014, pp. 744–6). Perhaps Quine simply gave modal realism
the same ‘incredulous stare’ that so many other analytic metaphysicians have
given it (Lewis 1986: 133–5). Moreover, Quine encountered modal realism at
a stage of his career when he was not disposed to accept radical revisions of his
views from outside; publicly retracting his well-entrenched signature scepticism
about quantified modal logic would have been a bitter pill to swallow. He showed
a similar lack of interest even in the purely technical development of the model
theory of modal logic, and in particular of quantified modal logic, by Kripke and
others, which is a piece of regular mathematics, no more vulnerable to Quine’s
concerns about intelligibility than any other piece of mathematics.
The example of Quine is a salutary reminder that the analytic tradition, as
normally understood, has never been a metaphysics-free zone. Before Quine,
Bertrand Russell was a major figure in the analytic tradition blatantly engaged
in metaphysical theorizing. Indeed, F.H. Bradley may well be right that critiques
of metaphysics themselves depend on contentious metaphysical assumptions
(Bradley 1893, pp. 1–2). Nevertheless, the role and status of metaphysics have
changed in significant ways over the history of the analytic tradition, and our
concern is with those ways.
Lewis’s case for his modal realism itself evolved over time. In the original
1968 paper, the emphasis is on the relation between modal and non-modal
languages, and the clarity to be achieved in modal logic by translating the
modal language into the non-modal language of Lewis’s counterpart theory
(the precursor of his modal realism). His postulates for counterpart theory
are used to validate elementary principles of modal logic, but they also clarify
the metaphysical picture. For instance, the postulate that nothing is in two
worlds has the advantage, according to Lewis, that it answers the question of
the identity of individuals across possible worlds at a stroke, with a uniform
negative. That is his answer to Quine’s complaints about the obscurity of trans-
world individuation (he cites Quine 1960, p. 245). By the time he wrote what
became the canonical case for modal realism, his book On the Plurality of Worlds
(Lewis 1986), based largely on his 1984 John Locke lectures at Oxford, Lewis’s
perspective had changed. He talks much less about linguistic matters, and
much more about the abductive advantages of modal realism as a theoretical
framework for explaining a variety of phenomena, many of them non-
linguistic. Faced with some objections to specific translation schemes between
the language of quantified modal logic (in effect, a formalization of ordinary
modal language) and the language of counterpart theory, he tells us not to worry
about the details, but instead to abandon the language of quantified modal logic
and do our metaphysical theorizing directly in the more perspicuous language
of counterpart theory (Lewis 1986, pp. 12–13). Clearly, from 1968 to 1986 the
balance of power in Lewis’s work swung towards metaphysics and away from the
philosophy of language.
Writing in 1981, Lewis described ‘a reasonable goal for a philosopher’ as

bringing one’s opinions into stable equilibrium. The trouble with losing ‘our
moorings in everyday common sense’, according to him, is not that common
sense is infallible but that we do not achieve stable equilibrium, since we keep
reverting to something like our everyday opinions (Lewis 1983a, p. x). He
requires the equilibrium to be stable under theoretical reflection. In principle, this
is not radically different from a Quinean methodology for philosophy and science
together of adjustments to ease tensions in one’s web of beliefs (Quine 1951).
For Lewis, modal realism does better than other theories of modal metaphysics
with respect to stability under reflection, given our other beliefs. Of course, many
philosophers would insist that to adopt modal realism is to lose one’s moorings
in everyday common sense, but Lewis was adept at interpreting the putative
deliverances of everyday common sense in ways that made them consistent with
modal realism, often by postulating large measures of tacit contextual restriction
in the utterances that gave them voice. Although he admitted some disagreements
between modal realism and common sense, for instance on whether (speaking
unrestrictedly) there are talking donkeys, he managed to steer the disagreements
into abstruse areas that might plausibly be regarded as of low priority for common
sense. Whereas critics condemned Lewis for his extravagant departures from
common sense, he saw his modal realism as part of his defence of common sense
– just the way Berkeley saw his subjective idealism.
In practice, the process of bringing one’s opinions into stable equilibrium
will involve extensive reflection on what one’s opinions actually are. Since one’s
general beliefs are typically presented to one as expressed by sentences, whose
underlying semantic structure is not perfectly transparent, in reflecting on what
it is one believes one is drawn into semantic reflection on one’s own language –
or, as some would have it, reflection on one’s own conceptual system.
A natural comparison is between Lewis’s Quinean or at least post-Quinean
methodology and the methodology of Peter Strawson, Quine’s leading opponent
from the tradition of ordinary language philosophy. By the late 1950s, however,
‘ordinary language philosophy’ was no longer an apt phrase for what Strawson
was doing. He was working in a far more abstract and systematic way than that
phrase suggests. His concern was with very general structural features of ordinary
thought and talk, such as the distinction between subject and predicate, rather
than with the fine detail of ordinary usage. In 1959 he published Individuals,
subtitled An Essay in Descriptive Metaphysics, a major monograph widely felt
at the time to mark a turning-point in the rehabilitation of metaphysics within
analytic philosophy. Strawson contrasted descriptive metaphysics with the wilder
revisionary metaphysics, which despite its ‘partial vision’ is nevertheless useful
when ‘at the service of descriptive metaphysics’ (Strawson 1959, p. 9). Revisionary
metaphysics can help at the periphery of our thinking, but goes astray when it tries
to revise ‘a massive central core of human thinking which has no history’ because
‘there are categories and concepts which, in their most fundamental character,
change not at all’ (ibid., p. 10). What descriptive metaphysics is supposed to
describe are the conceptual connexions that constitute the structure of that
central core. In Strawson’s view, that structure is not hierarchical, as envisaged

by programmes of conceptual analysis with their non-circular definitions by
necessary and sufficient conditions. Rather, the descriptive metaphysician traces
conceptual interconnexions on the same level, going round in complex closed
curves: the exploration is horizontal rather than vertical.
How different in kind is Strawsonian metaphysics, which may revise the
margins but must only describe the core of ordinary thinking, from Lewisian
metaphysics, which must not lose its moorings in everyday common sense in its
attempt to steer one’s opinions into equilibrium? At this point, it may be worth
recalling that Lewis attended lectures by Strawson, amongst others, when he
spent the academic year 1959–60 as a visiting student from Swarthmore College
at Oxford, tutored by Iris Murdoch, a year that resulted in his momentous
decision to major in Philosophy rather than, as he had previously intended,
Chemistry. Of course, Strawson’s characterization of descriptive metaphysics
as the tracing of conceptual connexions relies on some form of the analytic-
synthetic distinction, which he had defended with his old teacher Paul Grice
against Quine’s massively influential critique (Quine 1951, Grice and Strawson
1956). Indeed, the complex closed curve of definitions that Quine traced from
‘analytic’ round to other semantic terms and back again in his attempt to
show that none of them could be satisfactorily explained was just the sort of
explanation the descriptive metaphysician sought. But on this issue Lewis sided
with his earlier teacher against his later one. Lewis’s first book concluded with
an attempt to rehabilitate analyticity as truth in all possible worlds (Lewis 1969,
p. 208), to Quine’s regret in the book’s Foreword (Quine 1969, p. xii). Although
both Strawson and Lewis accepted an analytic-synthetic distinction, in their
philosophical practice neither of them had much tendency to use it as the
sort of glib conversation-stopper it so often becomes in less resourceful hands.
Admittedly, Strawson was more prone than Lewis to characterize philosophical
questions as questions about words or concepts, or as questions about how we
must think about a subject matter, rather than about the subject matter itself.
But Lewis himself frequently went metalinguistic, as we have already seen, and
Strawson was quite willing to speak in a ground-level metaphysical idiom, as
with remarks such as ‘A person is not an embodied ego, but an ego might be a
disembodied person’ (Strawson 1959, p. 103). They both moved easily between
object-language and meta-language, as it were, depending on the argumentative
needs of the moment.Quine insisted on continuity between philosophy and
natural science in a way that Strawson did not. In this respect, Lewis was closer
to Quine than to Strawson. In practice, however, natural science played only
a very minor role in the metaphysics of all three philosophers. A good test is
their treatment of the dispute between the three-dimensionalist Aristotelian
metaphysics of enduring continuants and the four-dimensionalist metaphysics
of occurrents composed of successive time-slices, often associated with Einstein’s
theory of special relativity. Predictably, Strawson is a three-dimensionalist while
Quine and Lewis are four-dimensionalists. But what is striking is how little Quine
and Lewis made of special relativity in their cases for four-dimensionalism. In
Word and Object, Quine emphasized the advantages in logical smoothness of

treating space and time on a par (Quine 1960, pp. 170–2). He adds only as a
convenient afterthought that Einstein’s discovery of special relativity ‘leaves no
reasonable alternative to treating time as spacelike’, but then immediately points
out that the logical benefits of doing so ‘are independent of Einstein’s principle’
and adds, with references, ‘the idea of paraphrasing tensed sentences into terms
of eternal relations of things to times was clear enough before Einstein’ (ibid., p.
172). In On the Plurality of Worlds, Lewis’s case for four-dimensionalism does
not mention Einstein or depend on modern science at all; it just relies on rather
shaky old-fashioned purely metaphysical reasoning about temporary intrinsic
properties, such as shapes (Lewis 1986, pp. 202–5). Presumably, Lewis was not
convinced that special relativity really did leave no reasonable alternative to
treating time as spacelike, otherwise he would have mentioned such impressive
support for his conclusion. Since the specifics of Lewis’s argumentation fail to
take seriously an approach that treats temporal operators as explanatorily basic,
it is hard to avoid the impression that what was really decisive with him was a
somewhat Quinean preconception about the proper sort of language for doing
metaphysics in, something close to the language of mathematics. Of course,
glancing through the pages of On the Plurality of Worlds, one sees them filled
with formula-free English prose, just like the pages of Individuals. Nevertheless,
the systematicity at which Lewis aims is modelled on that of a scientific theory
articulated in a mathematical language. The systematicity at which Strawson
aims is different; it is that of a satisfying general account in English itself, or
some other natural language. The sort of formal logical smoothness that Quine
and Lewis valued so highly, Strawson regarded as a trap. In this less standard
sense, despite the similarities, and even in their metaphysics, Strawson remained
at heart an ordinary language philosopher, and Lewis at heart an ideal language
philosopher. Each of them was reluctant to disagree extensively with common
sense (or natural science, for that matter), but left some room for manoeuvre by
acknowledging a belt of revisable opinions on the periphery of common sense.
As a result of quite a subtle difference in intellectual values, they ended up with
radically different theories.
What if Strawson is right, in that Lewis’s views conflict with the unchanging
core of ordinary thought? Then those views fail by Lewis’s own criterion, since
no totality of his opinions that includes them will be stable under theoretical
reflection, for it will also include the unchanging core of ordinary thought. We
cannot expect to keep questions about the general methodological position in
a philosophical debate clinically isolated from questions about who is right and
who is wrong on the specific matters at issue within the debate (Williamson
2007, pp. 210–14).
We can refine our sense of the intellectual options by comparing both
Strawson and Lewis with Kripke. In the 1970s, Kripke and Lewis were often
paired as leaders of the ‘possible worlds revolution’. Kripke’s essentialism and his
defence of quantified modal logic were radical in relation to Quinean orthodoxy.
It took time for them to be understood as articulations of quite ordinary
ways of thinking, so not radical in relation to common sense. Of course, the

connection between essentialism and Aristotle already noted by Quine was a
clue, since Aristotle always had a strong claim to be the founder of common
sense philosophy. Strawson names Aristotle and Kant as the great descriptive
metaphysicians (1959, p. 9). Kripke’s arguments in Naming and Necessity tend
to be based on common sense examples, and he explicitly rejects Lewis’s modal
realism, in favour of a more deflationary conception of possible worlds as possible
states of affairs (Kripke 1972). His metaphysics is arguably not revisionary in
Strawson’s sense. This feature of it may have been obscured by his technical
achievements in quantified modal logic, based on the mathematical apparatus
of ‘possible worlds semantics’, even though it is in itself no more metaphysically
problematic than any other piece of mathematics (Williamson 2013, pp. 81–4).
Although Kripke’s metaphysics initially could seem more revisionary than
it really was, his methodology could initially seem more linguistic than it really
was. His titles yoked together semantic and metaphysical terms: Naming and
Necessity (1972) and Reference and Existence (2013, but the book of his John
Locke Lectures given at Oxford in 1973 under the same title). Of course, it
was Kripke who played the central role in distinguishing metaphysical from
epistemic or semantic modalities, through his famous examples of contingent
truths knowable a priori and of necessary truths knowable only a posteriori,
which made trouble for the then-popular Humean slogan ‘All necessity is verbal
necessity’. Even so, there was a diffuse but influential impression in the 1970s
that Kripke had somehow managed to derive apparently substantial metaphysical
conclusions about the specific essential properties of individuals and kinds, from
his semantic analysis of modal language, in particular his insight that names are
rigid designators (their designation remains constant while different possible
worlds are considered). Nathan Salmon published a detailed monograph
Reference and Essence (1982) to refute that impression. The title uses ‘and’ to
separate the semantic term from the metaphysical one, rather than to join them
together. The front cover showed a rabbit being pulled from a hat. Salmon
demonstrated that Kripke had relied on metaphysical premises to derive his
metaphysical conclusion (Kripke himself had not claimed otherwise). That did
not make Kripke’s argument merely question-begging, for the plausibility of the
premises could still be more immediate than the plausibility of the conclusion.
Salmon’s point was widely accepted. That contributed to an increasingly popular
conception of metaphysics as separate from the philosophy of language.
Just as it was a mistake to regard Kripke’s metaphysics as derived from
his semantics, it would be a mistake to regard them as simply orthogonal to
each other. For misconceptions in semantics often induce misconceptions in
metaphysics, by causing fallacious metaphysical arguments to be treated as
valid, so that coherent metaphysical views are incorrectly dismissed as confused
or inconsistent. Quine’s early critique of quantified modal logic is an example.
It was crucial to get straight about the semantics of modal language in order
to see one’s way through to defending metaphysical theses of essentialism.
Thus Kripke’s semantic theory of rigid designators was after all relevant to his
essentialist metaphysics, but its role was negative, in driving off arguments
against essentialism, not positive, in driving forward arguments for essentialism.
There is a more general moral here about the famous ‘linguistic turn’, a
phrase which has looked less and less appropriate as a description of mainstream
analytic philosophy over recent decades. Nevertheless, although analytic
philosophers are ceasing to regard their central questions as linguistic or even
conceptual in any sense that would distinguish them from questions asked in
other disciplines, the traces of the linguistic turn are not simply being erased. For
it has left a rich legacy of methodological sophistication. In testing the soundness
of arguments about non-linguistic matters, analytic philosophers regularly draw
on work in both semantics and pragmatics (Williamson 2007, pp. 46–7). Kripke’s
work on quantified modal logic is a good example with respect to semantics.
With respect to pragmatics, the prime exhibit is the work of Strawson’s
teacher, Paul Grice, on conversational implicature (1961, 1975). If you comment
after my lecture ‘Williamson was sober this afternoon’, you imply that I am
often drunk in the afternoon, even though that is not a precondition for the
truth of what you said: it is even true if I am a scrupulous teetotaller. Grice
developed a powerful theoretical apparatus for analysing such effects. Although
this work emerged from within the Oxford of ordinary language philosophy,
it made an important contribution to undermining the methodology of such
philosophy. For ordinary language philosophy involved a focus on ‘what we
would say’ in various conversational contexts. By analysing the diversity of
reasons for which an utterance might be conversationally inappropriate, Grice
demonstrated the limitations to what can be concluded from such data. But
his theory of conversation was not just a factor in the implosion of ordinary
language philosophy; it has a far more lasting and positive value. It is massively
cited by linguists, because it is the starting-point for much contemporary work
in pragmatics. But it also continues to play a vital negative role in contemporary
analytic philosophy.
Analytic epistemology provides a good case study of the philosophical
application of Gricean pragmatics outside the philosophy of language. Analytic
epistemologists today typically regard the object of their study as knowing (or
justified believing) itself, as opposed to the corresponding words or concepts.
In reflecting on knowledge or justified belief, they work through example after
example of epistemologically suggestive situations, often of quite everyday
sorts. In determining how to describe such situations, they frequently have to
ask themselves whether a proposed description is false or, by contrast, true but
conversationally misleading because it has a false conversational implicature.
They use Grice’s theory of conversation to filter out contaminated data. They
also have to engage with semantics as well as pragmatics, since some of the
leading contender theories are contextualist, in the sense that they postulate
shifts in the reference of epistemic terms according to the context in which
they are used. Despite all that, the epistemologists’ underlying object of study
is knowing itself, not the verb ‘to know’ or the concept of knowing. They sound
like ordinary language philosophers, and in a loose enough sense they are
ordinary language philosophers, even when ordinary language plays no special

role in their epistemological aims (Hawthorne 2004 and Stanley 2005 are good
examples). In that respect, ordinary language has returned to its origins. For the
classic manifesto for ordinary language philosophy, at least in its Oxford form,
appeared in J.L. Austin’s paper ‘A plea for excuses’ (1956–7). But much of Austin’s
discussion of philosophical method there appears strongly influenced by similar
comments in the work of John Cook Wilson, Wykeham Professor of Logic at
Oxford from 1889 until his death in 1915, founder of an Oxford tradition of
realist metaphysics and knowledge-centred epistemology, and by no means
a linguistic philosopher as the term is usually understood. For instance, Cook
Wilson’s remark ‘Distinctions current in language can never be safely neglected’
(1926, p. 46) is echoed in Austin’s emphasis on the value of starting with
distinctions robust enough to have survived in ordinary language. Cook Wilson’s
ideas and writings still loomed large in the Oxford philosophy of Austin’s
student days, not least through the influence of his star pupil H. A. Prichard, the
White’s Professor of Moral Philosophy from 1927 to 1937, whose lectures Austin
attended as an undergraduate – despite Prichard’s attempt to ban him for asking
too many questions. Thus many contemporary analytic philosophers pay close
attention to linguistic subtleties, without treating their primary subject matter as
in any way linguistic.
The foregoing sketch suggests no easy moral, except that the closer one
looks at the history of anything, the less it lends itself to easy morals. But I can
add some colour to the picture, and another perspective on the transition, by
going back to my own experience of it at Oxford.
II
I arrived as an undergraduate at Oxford in 1973, to study Mathematics
and Philosophy. Logic played a central role in the course, to my lasting benefit;
its centrality is relevant to the point of view from which I observed the scene. I
received my undergraduate degree in 1976 and began to study for a doctorate.
Originally I hoped to formalize Leibniz’s principle of sufficient reason, but I soon
switched to Karl Popper’s idea of verisimilitude, on which science can progress
through a succession of theories that get closer and closer to the truth without ever
quite reaching it. I left Oxford in 1980, to start my first proper job, as a lecturer in
philosophy at Trinity College Dublin, and received my doctorate in 1981.
In 1973, the two senior professors of theoretical philosophy at Oxford were
A. J. (Freddie) Ayer, Wykeham Professor of Logic from 1959 to 1978, and P. F.
(Peter) Strawson, Waynflete Professor of Metaphysics from 1968 to 1987. Austin
had died in 1960 and Prior in 1969, both prematurely; Grice had left for Berkeley
in 1967. One could schematically associate Ayer and Strawson with the two main
strands of mid-century analytic philosophy: logical positivism and ordinary
language philosophy respectively. Ayer had studied with the Vienna Circle
and his first book, Language, Truth and Logic (1936) contained much logical
positivist doctrine, including a critique of metaphysics based on the verification
principle, although he traced his genealogy further back: ‘The views which are
put forward in this treatise derive from the doctrines of Bertrand Russell and
Wittgenstein, which are themselves the logical outcome of the empiricism of
Berkeley and David Hume’ (1936, p. 31). The book had been notorious for its
advocacy of expressivism about morality and religion. Its cheeky, provocative
style is conveyed by the title of the last chapter, ‘Solutions of outstanding
philosophical problems’. Whereas Ayer was a follower of Russell, Strawson’s most
famous article (1950) was a critique of Russell’s prize contribution to philosophy,
his theory of descriptions, and Strawson’s first book argued more generally that
such applications of modern logic did no justice to the subtleties of ordinary
language (1952). In live discussion, Ayer used rapid fire, Strawson elegant
rapier play. Ayer was better known to the general public, as a radio personality;
Strawson was more highly rated by professional philosophers, as more original.
Strawson had been a candidate for the Wykeham Chair of Logic; when Ayer was
elected, through the votes of the non-philosophers on the committee, Austin and
Ryle resigned in protest. Asked that evening by a colleague whether he felt very
disappointed not to have been elected, Strawson replied ‘Not disappointed, just
unappointed’.
By 1973, it was no longer strictly appropriate to classify Ayer as a logical
positivist, or Strawson as an ordinary language philosopher. It was closer to the
mark to describe Ayer as a Humean, and Strawson as a Kantian: the contrast
between them was no less marked for that. Strawson’s development into a
systematic metaphysician has already been noted. Ayer had retreated from
his early radicalism, including the verificationist critique of metaphysics. He
commented that the trouble with Language, Truth and Logic was that all its main
doctrines were false. In 1976, on the fortieth anniversary of its first publication, he
gave a series of lectures about what remained of his original view. In the book, he
had quoted as an example of an unverifiable pseudo-proposition of metaphysics
‘the Absolute enters into, but is itself incapable of, evolution and progress’, which
he describes as ‘A remark taken at random from Appearance and Reality, by F. H.
Bradley’ (1936, p. 36). In his lecture, he admitted that, far from having taken the
remark at random, he had searched through Appearance and Reality for hours
to find something suitably nonsensical-sounding. It is a salutary reminder of the
intelligibility by ordinary standards of much metaphysical discourse –especially
when not torn out of context in the way he presented the passage from Bradley.
To younger philosophers in 1973, Ayer appeared quite old-fashioned
philosophically. So too, though to a lesser extent, did Strawson. The underlying
reason was in large part their relation to modern formal logic in philosophy.
Officially, Ayer was for it and Strawson against it, but neither of them knew much
about it. They had received their philosophical education at a time when such
logic did not loom large in Oxford. Philosophers of that generation sometimes
referred to formal logic as ‘sums’, the primary school word for elementary
arithmetic. The effect of Ayer and Strawson’s lack of facility with modern formal
logic was that they were poorly placed to deal with the new wave of philosophy
of language sweeping across the Atlantic, led by Kripke and Lewis (who was at
least as much a philosopher of language as a metaphysican in the 1970s), and

other philosophers and linguists such as Donald Davidson, Hilary Putnam,
David Kaplan, Robert Stalnaker, Keith Donnellan, Richard Montague, and
Barbara Hall Partee. New wave philosophy of language involved the application
of formal semantics, based on modern logic, to natural languages.
Ayer had never had a detailed interest in the philosophy of language.
His resentment of the new wave focussed on Kripke’s case for the necessary a
posteriori and the contingent a priori, in effect because those categories violated
Hume’s supposedly exhaustive distinction between matters of fact (contingent
and a posteriori) and relations of ideas (necessary and a priori). An annual
ritual took place in Ayer’s ‘Informal Instruction’, his class open to all comers,
where a short presentation of some recently published work would kick off
the discussion. Many of the brightest graduate students attended, even though
they had joined the new wave, for Ayer was good at creating an atmosphere
conducive to discussion. But every year he would read a short paper purporting
to refute Kripke on the necessary a posteriori and the contingent a priori, in fact
based on exactly the confusions Kripke had done so much to clear up. When he
had finished, the graduate students would by implication plead with Ayer not to
misunderstand Kripke – in vain. Strawson was much more of a philosopher of
language than Ayer, but even his perception of new wave philosophy of language
was distorted by the old-fashioned lens of an exaggerated contrast between, in
effect, ordinary language philosophy attentive to speakers’ actual use of natural
language in all its complexity and ideal language philosophy trying to project
the simple logical structure of a formal language onto natural language, in
abstraction from its speakers, with Procrustean effect (Strawson 1971). What
he never properly appreciated was the new wave conception of the two projects
as mutually complementary rather than in competition, so that interpreting a
natural language in terms of a comparatively simple formal truth-conditional
semantics would make the best sense of the complexities of speakers’ actual use
of the language (Lewis 1975). That Strawson’s criticisms of new-wave philosophy
of language were widely felt to miss the point contributed to his looking like a
figure from the past too.
In effect, new wave philosophy of language achieved a surprising
reconciliation between elements from the two main competing strands of mid-
twentieth century analytic philosophy, logical positivism and ordinary language
philosophy. From logical positivism it took the rigorous use of formal languages
with precisely and systematically described syntax and semantics, as found in
modern logic, to model meaning. From ordinary language philosophy it took
most of its data about use to be explained, as well as ideas about the nature of
the relation between meaning and use, in order to bring the formal models to
bear on the data. An encouraging precedent was Noam Chomsky’s success in
explaining subtle, apparently messy complexities in the surface syntax of English
in terms of formal models of a postulated underlying deep structure (Chomsky
1957). What new wave philosophers of language hoped to do for the semantics
of natural languages seemed analogous to what Chomsky and others were
already doing for syntax – despite Chomsky’s own scepticism about the scientific
status of semantics. Indeed, it was natural to expect a tight relation between the
semantic and syntactic structure of an expression, at least at the level of deep
structure or logical form. For it was a fundamental tenet of new wave philosophy
of language, coming through Carnap from Gottlob Frege, that the semantics
must be compositional, in the sense that the meaning of a complex expression is
determined by the meanings of its constituents; how else to explain our ability
to understand sentences we have never previously encountered, if made up of
familiar words in familiar types of combination? The initial hypothesis must
surely be that the requisite semantic articulation of sentences into their semantic
constituents matches their syntactic articulation into syntactic constituents at
some deep enough level. The compositionality constraint exerted a powerful
force in the direction of systematicity. In practice, the only semantic theories to
exhibit (rather than merely claim) such a compositional structure were those for
formally specified languages. Without such a formal semantics, a philosophy of
language looked badly undeveloped to new wavers. It was partly for this reason
that Austin was barely mentioned in Oxford philosophy of language by 1973,
since he had no formal semantics to offer. The same went for his protégé and in
some respects intellectual heir John Searle, whose major work Speech Acts looked
out of date from an Oxford perspective almost as soon as it was published in
1969. Although Austin and Searle continued to exert a significant influence, it
was mainly outside the new wave.
New wave semantics came in two main varieties, though methodologically
the differences between them were minor compared to their shared differences
from their predecessors. One variety was possible worlds semantics, which
went back to Carnap and by 1973 was associated with philosophers of language
such as Kripke, Lewis, Kaplan, Stalnaker, and Montague. In ‘English as a formal
language’ (1970), ‘The proper treatment of quantification in ordinary English’
(1973), and other papers, Montague showed how it could provide a rigorously
working compositional semantics for large fragments of a natural language. His
work had a major influence on Barbara Hall Partee and has been seminal for a
major tradition of intensional semantics as a branch of linguistics. The other main
variety of new wave semantics was extensional, under the influence of Quine.
Chastened by his scepticism about meanings, it approached the semantic realm
less directly, through theorizing explicitly about reference and truth rather than
meaning itself. Nevertheless, its emphasis on the constraint of compositionality
was just as strong. Formally, it took inspiration from Tarski’s theory of truth. Its
main proponent was Donald Davidson (1967a). It too had a significant impact
on linguistics, most notably through Davidson’s semantics for verbs and adverbs
in natural language, which postulated tacit quantifiers over events (Davidson
1967b). A leading Davidsonian linguist was James Higginbotham.
I first encountered new wave philosophy of language in my first term as an
undergraduate, when my tutor encouraged me to attend the John Locke lectures,
to be given by the rising young star of American philosophy, Saul Kripke. I was
hugely impressed by his clarity, informal rigour, pointed examples, common
sense, and humour. The lectures were mostly non-technical, but one both sensed
and independently knew of his easy technical mastery of the subject. Although
I did not think of it this way at the time, Kripke combined and reconciled the
virtues of ideal language philosophy with those of ordinary language philosophy.
At that stage, of course, I knew very little of the background in the philosophy
of language to what he was saying, and was in no position to follow everything
that went on in the lectures and the discussion that followed them. Nevertheless,
Kripke became the nearest I had to a model of how to do philosophy.
Although Kripke’s work was widely discussed in Oxford at that time,
especially by younger philosophers, the dominant variety in town of new wave
semantics was extensional rather than intensional. Davidson had given the John
Locke lectures in 1970. The two most admired young theoretical philosophers in
Oxford, Gareth Evans and John McDowell, somehow combined Davidson with
Frege as the packed audiences at their joint classes looked on. It was the moment
of the ‘Davidsonic boom’ – although people tended to utter the Tarskian mantra
‘The sentence „Snow is white” is true in English if and only if snow is white’ in
a rather slow and quiet voice. If you wanted to write on the philosophy of X,
whatever X was, then you were supposed to start by writing a truth theory for a
language for talking about X. I found the atmosphere of reverence for Davidson
unhealthy, not to say sickening. It was fine for him to choose a speculative and
controversial extensionalist starting point for his programme for the philosophy
of language, and to take it as far as he could from there, but the project cried
out to be undertaken in a scientific spirit, as a test of its assumptions. Instead,
they were treated – especially by those lower in the hierarchy – as dogmas of
mysterious but compelling power, an attitude encouraged by Davidson’s elliptical
and slightly evasive style. It was recognized that Davidson’s programme had to
meet the challenge of providing a compositional semantics for various apparently
non-extensional constructions in natural language, for ascribing beliefs and
desires or possibility and necessity, for example, but the discussion of alternative
proposals muffled the issues with a lack of openness and clarity about the rules
of the game. Philosophers, some with only a rather tenuous grasp of technical
matters, would invoke obscurely motivated technical constraints – a ban on
substitutional quantification, say, or a requirement of finite axiomatizability
– to exclude rival hypotheses. Kripke’s critique of Davidsonians’ objections to
substitutional quantification (1976) has been condemned as cruel, but to me it
came as a breath of fresh air; I can attest to the presence at the time of the sort
of atmosphere about which he complained.2 By contrast, intensional semantics
seemed to be conducted in a more open, scientific spirit, though Davidsonians
objected, darkly, to its possible worlds as creatures of darkness.
2 My first article to be accepted for publication, though not my first to be published, protested
against the Davidsonian dogma that theories of truth qua theories of meaning had to be
finitely axiomatizable (Williamson 1984).
Davidsonians did not expect the philosophy of language to be independent

of metaphysics. Davidson (1977) explicitly motivated an ontology of events by a
Quine-inspired principle of charity in interpretation, through the semantics of
adverbs. This was not so different from Lewis’s original motivation of his ontology
of possible worlds, in effect by the principle of charity, through the semantics
of modal operators. But Davidson gave his metaphysics a turn reminiscent of
Strawson, with a transcendental argument to show that it was not really possible
to think differently (Davidson 1973–74). As so often with transcendental
arguments, it turns out to depend on concealed verificationist assumptions (see
the remarks below on Wittgenstein’s Private Language Argument). More recent
metaphysics has been much less tempted by transcendental arguments.
Amongst the important features shared by the extensional and intensional
varieties of new wave semantics was truth-conditionality: they both treated the
meaning of a declarative sentence as in some sense the condition for it to be
true. By contrast, the senior Oxford philosopher properly to engage new wave
philosophy of language, Michael Dummett, opposed such truth-conditional
semantics in favour of assertibility-conditional semantics, which treated the
meaning of a declarative sentence as the condition for it to be assertible, or
verifiable, rather than true. Assertibility-conditional semantics was inspired
by the proof-conditional semantics for mathematical language developed
by Heyting, Prawitz, and other intuitionists, which equated the meaning of a
sentence in the language of mathematics with the condition for something to be
a proof of it. The plan was to generalize this semantics to the whole of language
by treating mathematical proof as a special case of verification. This may be seen
as a descendant of a logical positivist conception of the meaning of a sentence as
its method of verification, although Dummett did not present it as such.
Unlike the logical positivists, Dummett saw the subversive threat that
verification-centred semantics posed to classical logic, given the compositionality
constraint which he accepted, in line with new wave philosophy of language. For
example, the natural compositional semantic clause for disjunctive sentences of
the form ‘A or B’ says that ‘A or B’ is verified if and only if either ‘A’ is verified or
‘B’ is verified. But that makes immediate trouble for the classical law of excluded
middle, ‘A or not A’. For the special case of the semantic clause where ‘B’ = ‘not
A’ says that ‘A or not A’ is verified if and only if either ‘A’ is verified or ‘not A’ is
verified. But often we cannot verify a sentence and cannot verify its negation.
For instance, we cannot verify ‘Napoleon had an even number of hairs at his
death’, nor can we verify ‘Napoleon had an odd number of hairs at his death’.
Thus, by the semantic clause, we cannot verify ‘Napoleon had an odd or even
number of hairs at his death’.
In these ways, Dummett saw the philosophy of language as calling into
question a realist metaphysical conception of reality as how it is quite independently
of our capacity to find out how it is, and pointing towards an alternative anti-
realist metaphysics. In his view, the role of the philosophy of language here is not
merely evidential, to give us reasons to believe one metaphysical theory and not
another. Rather, he saw alternative theories of meaning as giving something like
the cash-value of alternative metaphysical pictures. Methodologically, he proposed

to replace futile quarrels between metaphysical pictures by comparisons between
the corresponding theories of meaning as the only scientific way to resolve the
issue. For him, metaphysical disputes are not senseless; nor are they what they
seem, since they are implicitly disputes in the philosophy of language. One might
compare Dummett’s understanding of metaphysics with Strawson’s. For both of
them, metaphysical questions turn into questions about the structure and limits of
coherent thought, to be answered by systematic inquiry. But Dummett was more
open to revisionary metaphysics than Strawson was, because he took seriously the
danger of fundamental incoherence in our current ways of thinking (as with our
acceptance of the law of excluded middle), which Strawson did not. Dummett
also differed from Strawson in regarding modern logic as a decisive advance over
its predecessors, including its capacity to provide formal methods and model
formal languages for use in the philosophy of language: he was adept himself in
such methods. In that sense too, he was a new wave philosopher of language.
Dummett laid out his programme early on in his career (1959). The year
Kripke gave the John Locke lectures, Dummett brought out his first magnum
opus, Frege: Philosophy of Language (1973), in which he engaged creatively with
Frege to develop his own views. Central to Oxford philosophy in my student days
was the dispute between, on one side, realism and truth-conditional semantics,
represented by the Davidsonians, and, on the other side, anti-realism and
assertibility-conditional semantics, represented by Dummett. My sympathies
were strongly with realism, though not with Davidsonianism. Dummett
supervised me for the last year of my doctoral studies (1979–80), at the start
of his period as Wykeham Professor of Logic (1979–92), the first holder of that
chair with a deep knowledge of modern logic. He was remarkably tolerant of the
strident realism of my thesis, which effectively presupposed the futility of his
life’s work and pursued other issues from that starting-point. I cannot resist a
couple of memories from that period.
When I told other philosophers at Oxford that I was working on the idea
of approximation to the truth as applied to scientific theories, their reaction was
always to ask „Is that something to do with vagueness?” For vagueness was a big
issue in Oxford then, being conceived as a major challenge to realism, truth-
conditional semantics, and other forms of orthodoxy (Dummett 1975, Fine
1975, Wright 1975). I always found that reaction annoying, because I thought
it betrayed a myopic obsession with the philosophy of language. I would reply
that my thesis had nothing at all to do with vagueness, pointing out that, of two
perfectly precise but false scientific theories, one may be a better approximation
to the truth than the other (I also enjoyed shocking people by saying that I found
Popper more interesting than Davidson). On the narrow issue I was right, but
my later trajectory suggests that my interlocutors were not completely wrong
about the direction of my interests (Williamson 1994). Indeed, one of my main
reasons for later working on vagueness was that it was generally regarded as a
paradigm of a phenomenon in need of anti-realist treatment. I wanted to strike
at what was supposed to be the safest fortress of anti-realism.
Another memory comes from a supervision with Dummett. We were

discussing an argument that I thought one of the best in my thesis, and he
thought one of the worst; it later became the kernel of the only publication that
emerged from my doctoral studies (Williamson 1988). After a while, Dummett
reflected and said „The difference between us is that you think that inference
to the best explanation is a legitimate method of argument in philosophy, and I
don’t”. I realized that his characterization of the difference was right, although I
was a little shocked at his outright rejection of inference to the best explanation
in philosophy. His view was something like this: the deep philosophical issue will
be about which of the theories that yield the putative explanations is so much
as meaningful; that issue must be settled first before we can judge the value of
those putative explanations; but once it has been settled, nothing much is left
for inference to the best explanation to do. I still favour inference to the best
explanation and an abductive methodology in philosophy (Williamson 2013, pp.
423–9). Indeed, it is hard to see how the kind of positive, systematic, general
theory that Dummett sought in the philosophy of language could be established
by any other means. He was optimistic about the long-run prospects of settling
philosophical issues in a decisive, systematic way, just as he took to happen in
science (Dummett 1978). But if one tries to establish the meaningfulness of a
theory by an argument more decisive than abduction, won’t one have to first
establish the meaningfulness of that argument by a further argument more
decisive than abduction? There starts an infinite regress. Dummett’s dislike
of inference to the best explanation may help to explain why his discussion of
assertibility-conditional semantics never really got beyond the programmatic
stage to the nitty-gritty of properly developing models of such semantics for
non-trivial fragments of non-mathematical language. Even if such models had
worked well (a tall order), they would at best have provided him with some sort of
abductive argument in favour of his programme, whereas he wanted something
more decisive. In the long run, the failure of his programme to develop such
working models has been a major reason for its marginalization, especially when
combined with its radically revisionary and implausible consequences for logic
and metaphysics.
Dummett presented his views on the relationship between the philosophy
of language and metaphysics in his William James Lectures at Harvard in 1976,
the same year that Hilary Putnam gave the John Locke Lectures at Oxford. A
book soon grew out of Putnam’s lectures, showing signs of Dummett’s influence
in some rather unfortunate arguments against something called ‘metaphysical
realism’ (Putnam 1978). But it was not long before the anti-realist Dummett was
replaced by the realist (if not metaphysical realist) J.L. Austin as the main Oxford
presence in Putnam’s work (Putnam 1994). Dummett’s lectures took much
longer to appear in print, as The Logical Basis of Metaphysics (Dummett 1991).
For a key work of a major philosopher, it had comparatively little impact. One
problem was that its discussion of proof theory as the foundation of semantics
was informal, often elliptical, digressive, or vague, with philosophical and purely
technical matters all mixed together, making it excessively and unnecessarily hard
for logicians to extract and perhaps answer the purely technical questions raised.
Moreover, the generalization of the semantics to non-mathematical language
still remained at a tentative, programmatic stage, not conducive to applications
in linguistics. A more general problem for Dummett was that by 1991 the
philosophical zeitgeist was even less receptive to anything like Dummett’s
programme than it had been earlier. He did not engage with the new paradigms
of metaphysics, such as the work of David Lewis. There was no obvious way to
interpret the new metaphysical theories as picturesque guises for views in the
theory of meaning, nor did the new generation of metaphysicians wish to do
so. Metaphysics itself had grown in self-confidence and felt no need to present
itself as anything else. Incidentally, despite its title, my own book Modal Logic as
Metaphysics (2013) is very far from a return to Dummett’s understanding of the
relationship between logic and metaphysics. As a first approximation, ‘logic’ in
Dummett’s title means something like ‘philosophical reflection on the meaning
of the logical constants’, while in mine it means ‘generalizing about the world in
terms just of the logical constants’. For Dummett, logic is metalinguistic, for me
it is not.
By the 1990s, few readers felt in danger of being compelled, against their
wills, by Dummett’s convoluted arguments, even when they understood them.
One of several reasons was that he discussed the mind in ways that still carried
behaviourist baggage from a philosophy of mind widely rejected by younger
generations since the collapse of behaviourist psychology and its replacement
by cognitive psychology in the 1960s. Such baggage is detectable in remarks
like this about knowledge of meaning: ‘we should [...] not be content with
saying what is known, without saying what it is to have that knowledge, that
is, how it is manifested by one who has it’ (Dummett 1991, pp. 104–5), where
the manifestation is in observable behaviour. In this respect, Dummett can be
compared to Quine, who was influenced by his Harvard colleague Skinner’s
behaviourism in psychology. One might think that behaviourism about language
had been outdated since the publication of Chomsky’s famously destructive
review of Skinner’s Linguistic Behavior (1959), but digestion can be a slow process.
I remember sophisticated young philosophers of language at Oxford in the late
1970s still talking of children learning their native language by being ‘drilled’
in it by adults. In Dummett’s case, his behaviourist tendencies came from his
reading of Wittgenstein rather than Skinner, and were correspondingly subtler
and less eliminativist than Quine’s. Nevertheless, the differences should not be
exaggerated. An anecdote from late in Dummett’s career: A group of younger
Oxford philosophers were discussing what he meant, in the great man’s silent
presence. Suggestion after suggestion was rejected because it would attribute
to him ‘crude old-fashioned behaviourism’. Eventually, someone turned to him
and asked ‘So what is your view, Michael?’. Dummett replied ‘I think it’s the one
you’ve been calling „crude old-fashioned behaviourism”‘.
For Dummett, as for other British philosophers of his generation,
Wittgenstein’s central contribution to the philosophy of mind was his Private
Language Argument. Its interpretation was disputed, but it was widely supposed
to show something very deep about the need for talk about mental states to
involve observable criteria (in some sense) for attributing them to others. The
putative insight had widespread repercussions for the philosophy of language,
concerning not just the semantics of mental state ascriptions but the nature
of the mental states in play for speakers and hearers of any speech act, and in
particular the nature of understanding. Dummett’s preference for assertibility-
conditions over truth-conditions in the theory of meaning was rooted in the
close linkage of assertibility-conditions to the observable use of the language,
which realist truth-conditions lacked, since users of the language might have no
idea whether they obtained. He combined this Wittgenstein-inspired focus on
use with a Frege-inspired insistence on the need for a systematic, compositional
theory of meaning, modelled on the semantics of a formal language. In thus
uniting elements of the two previous traditions of analytic philosophy, ordinary
language philosophy and ideal language philosophy, Dummett resembled the
younger new wave philosophers of language, although his selection of elements
to combine differed from theirs.
At this point something must be said more generally about the influence of
Wittgenstein on British philosophy in the period under discussion – as has often
been remarked, his influence in North America was never as great as in Europe,
one reason being the greater sway of naturalism or scientism in North America,
led by Quine and others. In the case of Dummett, since he was a student at
Oxford in 1950 when Wittgenstein spent some time there living in Elizabeth
Anscombe’s house, future historians might wonder whether there was face-to-
face influence. It is therefore worth recounting the story Dummett liked to tell
about his only meeting with Wittgenstein. Dummett was going for a tutorial at
Anscombe’s house. She kept the door unlocked. As was the practice, Dummett
went in, and sat down to await her summons. An elderly man in a dressing-gown
came downstairs and asked ‘Where’s the milk?’; Dummett replied ‘Don’t ask me’.
That was the extent of his conversation with Wittgenstein. What mattered instead
was Anscombe’s mediating role. She was probably the strongest transmitter of
Wittgenstein’s influence at Oxford until she left in 1970 to take up a chair at
Cambridge, although of course she was always a fiercely independent-minded
philosopher in her own right. Other Oxford ordinary language philosophers
such as Ryle, Austin, and Grice were not moulded by Wittgenstein, and the
number of card-carrying Wittgensteinians at Oxford was never very high.
Nevertheless, his influence was still pervasive when I was a student in the 1970s.
Gordon Baker and Peter Hacker, guardians of the flame, had a large following
amongst graduate students. They were later to have an ill-tempered dispute
about the value of Frege’s philosophy with Dummett: when their disparaging
book on Frege was published, Dummett organized an emergency series of
graduate classes to denounce it (Baker and Hacker 1984, Dummett 1984). It is
easy to list many Oxford philosophers of the time whose work showed significant
Wittgensteinian influence to varying extents, even though it would be crass to
classify them simply as Wittgensteinians: Dummett, Strawson, Philippa Foot, Iris
Murdoch, David Pears, Anthony Kenny, of a younger generation John McDowell
and Crispin Wright, and so on. What may be less obvious is how wary even
those who barely mentioned him were of plainly saying that he was wrong about
something. One knew that doing so incurred the automatic charge of shallow
misinterpretation. It was best to step quietly around, and let sleeping dogs lie.
Wittgenstein’s main influence at that time was through his later work,
although few of those under the influence imitated his style of philosophizing in
that work. Most engaged in overt theorizing of a more or less systematic kind. The
citadel was the Private Language Argument, from which he exerted his power over
the philosophy of mind and the philosophy of language. The growing external
threat to that power from cognitive psychology was surprisingly little felt in 1970s
British philosophy. But there was also an internal threat. For how exactly was the
Private Language Argument supposed to work? Wittgenstein’s presentation was
notoriously Delphic. The simplest and clearest reconstructions had the argument
rest on a verificationist premise to the effect that one couldn’t be in a mental state
unless some independent check was possible on whether one was in that state.
But it was generally agreed that if the argument rested on a verificationist premise
then it was not compelling, because verificationism could not just be assumed
without argument. Defenders of the argument insisted that it worked without such
a premise, but could not satisfactorily explain how (a similarity with Davidson’s
transcendental argument mentioned above). Wittgenstein’s citadel was in danger
from within; his power was waning as a result. At this point an unlikely would-be
rescuer arrived: Saul Kripke. In lectures from 1976 onwards, and in his book on the
Private Language Argument and the associated considerations on rule-following
(Kripke 1982), he offered a conjectural interpretation of the argument that was
clearly non-verificationist and, if not compelling, at least powerful. The question
was: did it fit what Wittgenstein meant? The consensus amongst Wittgensteinians
was that it did not, and as a matter of historical scholarship they may well have
been right. But they seemed not to realize that in taking the negative attitude
they did, they were also rejecting their last chance to avoid marginalization from
the philosophical mainstream. The power of Wittgenstein’s name resumed its
decline. As for Kripke’s argument in its own right, it inadvertently gave the new
metaphysics an opportunity to spread its influence. For Kripke’s argument took
the form of a sceptical paradox, to which Kripke offered a rather unclear and
unattractive radically sceptical solution. By contrast, David Lewis offered a clearer
and more attractive non-sceptical solution, by means of a highly metaphysical
distinction between objectively natural and objectively non-natural properties
(Lewis 1983b). It was something like Lewis’s solution, not Kripke’s, that was
widely accepted.
Here are two snapshots of the decline in Wittgenstein’s standing. The first
is of a meeting in about 1994 of the ‘Tuesday group’, originally founded by
Ayer on his return to Oxford in 1959 as a counterweight to Austin’s Saturday
morning meetings. Susan Hurley read a carefully reasoned paper against the
Private Language Argument to an audience that included many leading Oxford
philosophers. The audience divided by age. Roughly, those over fifty did not take
the possibility seriously that Wittgenstein’s argument was fundamentally flawed,
although they also did not explain how it worked or what it showed; those under
fifty were more sympathetic to Hurley’s objections. The second snapshot is of

a large graduate class on philosophical logic shortly after my return to Oxford
in 2000. One student kept pressing the Wittgensteinian line that contradictions
are meaningless rather than false. I kept giving the standard responses, that
contradictions have true negations while the negation of what is meaningless
is itself meaningless so not true, that the compositional semantics generates
meanings even for contradictions, and so on, whose effect was merely to elicit
variations on the same theme that did not meet the objection. Eventually I
became exasperated and said ‘Maybe Wittgenstein was just wrong; it wouldn’t be
the first time’. There was a collective gasp of shock. I have never again witnessed
such a reaction when Wittgenstein’s name was taken lightly.
Of course, the flame is kept alive by surviving groups of Old Believers.
Some others, more willing to believe that there has been progress in philosophy
since 1970, still find value in engaging with Wittgenstein’s work. Nevertheless,
his influenced has declined drastically over the past forty years. No doubt that
could be roughly measured by his proportion of citations in journals. But what
strikes me most forcefully is that the fear factor has gone. As a test of authority,
of intellectual or other kinds, admiration tells less than fear. In the 1970s,
even non-Wittgensteinian philosophers were often afraid to speak out against
Wittgenstein. They are so no longer. Another philosopher who has ceased to
elicit the fear factor is Quine. Originally, he was frightening because few could
match his skill with the weapons of formal logic in philosophical debate. By the
1970s that was no longer so, but philosophers were still very nervous of relying
on ordinary semantic notions such as synonymy, because they were afraid of
being caught out by Quine’s argument for the indeterminacy of translation. That
fear too gradually evaporated in the 1970s, as Quine’s behaviourist assumptions
fell into disrepute.
Having said so much about the Oxford philosophical scene in the 1970s,
I should continue the story into the 1980s and 90s. It was not at all a linear
extrapolation from the 1970s. Strikingly, new wave philosophy of language
receded fast (though it proved temporarily) in Oxford, less so elsewhere. One
reason was tragically extrinsic: Gareth Evans died at the age of 34 in 1980. With
him, the new wave in Oxford lost much of its technical panache, and detailed
work in semantics dropped off. Although James Higginbotham was Professor of
General Linguistics from 1993 to 2000, not much else was going on at Oxford in
Davidsonian semantics for him to engage with. The Davidsonic boom had come
to an odd end, morphing into moral philosophy in the work of John McDowell,
David Wiggins (Wykeham Professor of Logic from 1994 to 2000, in succession
to Dummett), Mark Platts, and others. The transition was made through the
Davidsonian emphasis on the legitimacy of homophonic truth theories, in
which a word is used to state its own reference. Contrary to appearances, it is
not trivial that ‘round’ in English applies to all and only round things, because
one has that to learn in learning English. Semantic analysis cannot go on for
ever; eventually we reach semantic atoms, and switching to a non-homophonic
semantics achieves nothing to the purpose, because the aim of semantics
is not to write a textbook that one might read in order to learn the object-
language from scratch, but rather to say explicitly what its expressions mean in
a systematic, compositional way, to those who may already understand them
implicitly. Those in the intensionalist strand of new wave philosophy of language
had to resort to homophonic lexical semantics too. The Davidsonians realized
that, in particular, they could just as well give a homophonic semantics for moral
language too (Wiggins 1976). For instance, ‘evil’ in English applies to all and
only evil things. Nothing in their philosophy of language made it problematic
to give such an ostensibly out-and-out realist treatment of moral language. Nor
did it require any further semantic analysis of moral terms; they could be treated
as unanalysable. Thus Davidsonian philosophy of language found itself in the
unaccustomed role of providing a protective environment for Aristotelian moral
realism. By contrast, Dummett put much heavier explanatory demands on the
theory of meaning, perhaps too heavy to be satisfiable.
From the late 1970s onwards, Dummett also found himself fighting a
more global trend in analytic philosophy: a move away from the philosophy
of language towards the philosophy of mind. On his picture of the history of
philosophy, Descartes had made epistemology first philosophy, the engine for the
rest of philosophy, and then Frege had replaced epistemology by the philosophy
of language as first philosophy. Analytic philosophy was philosophy downstream
from that linguistic turn. But many analytic or ex-analytic philosophers were
starting, heretically, to treat the philosophy of mind as more fundamental than
the philosophy of language. Cognitive psychology made a far more interesting
and attractive conversation partner for philosophy than behaviourist psychology
had done, and had a natural interface directly with the philosophy of mind, for
instance in the theory of perception. Computer models of the internal workings of
the mind were also increasingly influential. Once again, many of the innovations
came from North America. As behaviourism lost its authority, Thomas Nagel
(1974) led the way in talking directly about conscious experience. Although
Daniel Dennett (1981) still showed some influence from his Oxford supervisor
Ryle, he engaged with psychology through the philosophy of mind, not the
philosophy of language. Jerry Fodor (1975) postulated a language of thought,
on the model of a computer’s machine code, but it was to be studied by the
methods of psychology and computer science, not those of linguistics. Moreover,
it was not a public language, whereas Dummett envisaged first philosophy as the
philosophy of public language, in line with the Private Language Argument.
In Oxford, the move into the philosophy of mind took a very specific
form, which Dummett had unintentionally facilitated. For him, much of Frege’s
achievement in the philosophy of language depended on his distinction between
sense and reference. Sense is individuated cognitively: two senses may present the
same reference in ways which count as different because they fail to render the
sameness of reference transparent to the thinker. The cognitive nature of sense
promised to make the connection Dummett wanted between the semantics of a
language and speakers’ use of that language. Thus ‘Hesperus’ and ‘Phosphorus’
differ in sense and use but not in reference. Dummett followed Frege in
making sense a level of linguistic meaning distinct from the level of reference.
Initially, this gave Fregean semantics a large head start over one-level referential
semantics, like that associated with Russell, in explaining linguistic phenomena

such as the apparent difference in truth-conditions between ‘Mary thinks that
Hesperus is bright’ and ‘Mary thinks that Phosphorus is bright’. However, new
wave philosophy of language in North America tutned against Frege, especially
at the level of public language. In particular, the semantic property of a name that
speakers share is its reference; as Kripke and even Frege emphasized, the name’s
cognitive connections may vary wildly from one speaker to another. Something
similar applies to terms whose reference depends on context: the linguistic
meaning of the phrase ‘that dog’ does not encode the rich cognitive connections
that it will have when used as a perceptual demonstrative by a particular speaker
on a particular occasion. Most of the younger Oxford philosophers of language
in the 1970s and 1980s followed Dummett in his Fregean sympathies. But, more
impressed than Dummett by the work of Kripke and other North American new
wave philosophers of language, they applied the sense-reference distinction only
at the level of individual users of the language, not at the level of the language
as a whole. If senses are cognitively individuated determinants of reference, a
proper name expresses different senses for different speakers, and a perceptual
demonstrative expresses different senses on different occasions even for the
same speaker. If senses are structured, much of that structure will be at the level
of thought and not at the level of language. ‘Sense’ was often glossed as ‘a way
of thinking of the referent’. This shift in focus from language to thought was
already visible in the work of Evans (1982). By Dummett’s standard, it meant
that Evans and the others who took that turn no longer even counted as analytic
philosophers.
Thus Dummett found himself fighting on the home front too, trying to
reassert the primacy of language over thought in philosophical method. Although
he was happy to regard philosophy as the study of thought – of what is thought,
not the act of thinking it – he insisted that the proper way for philosophers to
study thought was by studying it as expressed in public language, which the neo-
Fregean philosophers of thought no longer did. Perhaps they were not in direct
contravention of the Private Language Argument, because their senses could in
principle be shared. Nevertheless, from Dummett’s methodological perspective,
they had taken a step backwards, because the study of public language gave
philosophy the objective discipline it needed. To replace that discipline by the
objective discipline of experimental psychology would be, from his perspective,
to commit the disastrous error of psychologism, against which Frege had railed:
it would involve confusing what is thought with the act of thinking.
Dummett seemed to be fighting a losing battle. Globally, the centre of
gravity of analytic or post-analytic philosophy moved towards the philosophy
of mind in the 1980s. Logic and semantics suffered a significant loss of prestige:
graduate students became less convinced of the need, intellectual or professional,
to put in the hard work of learning them. Locally, neo-Fregean philosophers of
thought were taking over. For instance, Strawson was succeeded as Waynflete
Professor of Metaphysics by Christopher Peacocke, who held the post from 1988
to 2000. Senses became concepts (Peacocke 1992).
Since the 1980s, the philosophy of mind worldwide has continued to enjoy
a far more fruitful relationship with experimental philosophy than it did in
the heyday of behaviourism. However, it did not become first philosophy in
the way it had been expected to do. Nor did the philosophy of thought, which
anyway never solidified as a recognized branch of the subject. For instance,
developments in metaphysics have typically not been driven by anything in
the philosophy of mind. After all, with regained confidence in metaphysics, its
contemporary practitioners tend to see themselves as investigating the most
general and fundamental nature of a world in which human minds play only a
very minor role. Why should the philosophy of mind or the study of concepts
drive metaphysics any more than it drives physics? In principle, even if it cannot
contribute towards constructive metaphysical theory-building, it might help
towards understanding the folk metaphysical beliefs that may obstruct our
acceptance of the correct revisionary metaphysics. In practice, the philosophy
of mind and the study of concepts have had little impact on recent mainstream
metaphysics even in that modest negative way.
For the past several decades, no branch of philosophy has played the fully-
fledged role of first philosophy within analytic philosophy. To some extent,
that reflects the increasing specialization of academic research in general.
But it also concerns a change more specific to analytic philosophy (in a sense
broader than Dummett’s), in what philosophers take their subject matter to be.
As already noted, an increasingly prevalent, broadly realist attitude is that when
you are doing the philosophy of X, you are primarily interested in X itself, in
its most general and fundamental aspects, and only secondarily in the word
‘X’, or our concept of X, or our beliefs about X, or our knowledge of X. You
are not surreptitiously doing the philosophy of language or thought or mind or
knowledge. This reconception of the subject gives no branch of philosophy a
head start over the others.
However, the situation is more complicated than those simple formulations
suggest. For they might lead one to expect the philosophy of language to be
just one more branch of philosophy alongside all the others, the philosophy of
a phenomenon specific to humans and perhaps some other species scattered
here and there over the universe. It looked like that to some in the 1980s, and
taken in isolation it may still sometimes look like that. But, as already suggested,
the philosophy of language also plays a more general role throughout analytic
philosophy, in the evaluation of arguments. Of course, we do not need the
philosophy of language to determine whether an argument is deductively
valid in simple cases. But on almost any view of philosophy, it often involves
arguments with a subtle illusion of validity, and other arguments that are really
valid but need to be checked for such subtle illusions. The illusion may come
from confusions between entailments and presuppositions or conversational or
conventional implicatures, or from concealed shifts in context, or from lexical
or syntactic ambiguities, or from other linguistic complexities. Any discipline
that uses subtle, complex would-be deductive arguments in natural language
about abstract issues is liable to such illusions, and philosophy characteristically
uses arguments of that kind. That is of course not to say that it uses nothing
else, or that no other discipline uses them at all; nevertheless, philosophical

methodology past and present may put more weight on such arguments than
does the methodology of any other discipline. The shift from a deductive to
an abductive methodology makes less difference here than one might have
expected, because abduction involves the assessment of – amongst other factors
– the strength, explanatory power and consistency with the evidence of a theory,
which in turns depends on its deductive consequences. Thus simply using the
methods of analytic philosophy critically, by contemporary standards, takes
some sophistication in both semantics and pragmatics, irrespective of the subject
matter under philosophical investigation. That is a robust legacy from analytic
philosophy of language for all philosophy.
One day, perhaps, cognitive psychology will have developed to a point at
which it can be usefully deployed to locate likely trouble spots for philosophical
reasoning, for instance where framing effects may be exerting an undue influence.
Some ‘experimental philosophers’ believe that we have already reached that
point. However, perhaps with a few limited exceptions, it is doubtful that purely
psychological methods have yet reached an adequate level of discrimination to be
usefully applied in the way that linguistic methods already can be. Just to be told
that the order in which material is presented can influence our judgment is of
little help, since either we ignore the material or it is presented in some order or
other. For the time being, linguistics and the philosophy of language offer more
help than do psychology and the philosophy of mind when we check an alleged
deduction. In this limited respect, Dummett was right about the methodological
danger of assigning priority to thought over language, but not for the deep and
permanent reasons he envisaged.
For the evaluation of deductive arguments, the relevance of logic is even
more obvious than that of the philosophy of language. Of course, some would-be
deductive arguments in philosophy are cast in such seamlessly discursive form
that no extant logical theory is of much use in evaluating them. Nevertheless,
in most branches of contemporary analytic philosophy complex would-be
deductive arguments often are articulated with sufficient clarity for formal logical
skills to make a significant difference to the reliability with which their validity
is assessed. Thus logic makes an instrumental contribution to philosophy in
general similar in kind to that made by the philosophy of language, and perhaps
greater in degree.
The development of formal methods in recent philosophy has also extended
the scope for logic to make a more direct contribution to branches of philosophy
usually conceived as ‘other’ than logic. For instance, in epistemology, models of
epistemic logic enable us to work through the consequences of epistemological
claims in exactly described, appropriately simplified situations in a far more
rigorous and systematic way than would otherwise be available. Something
similar goes for decision theory too. The model-building methodology that has
proved so successful in the natural sciences can thereby be applied in philosophy
too, and provides new insights into old problems. In metaphysics, rival logics
often supply powerful structural cores to rival metaphysical theories: for instance,
a quantified modal logic is the structural core of any properly developed theory
of modal metaphysics. Although not all of modal metaphysics is usefully treated
as logic, a vital part of it is. Logic, far from displacing metaphysics as the logical
positivists hoped, is at its centre.
The history of philosophy makes a mockery of any limited vision of what
philosophy is. It has not followed the path laid out for it by the logical positivists,
nor that laid out by the ordinary language philosophy. Nor has i perhaps
with a few limited exceptions t become a branch of psychology, or of physics.
Nevertheless, under all the surface turbulence, it somehow manages to extract
the residue it needs from each changing fashion. Who knows where the cunning
of reason will take it next?
III
History is often said to be written by the winners. In the case of analytic
philosophy, however, there is a danger that history will be written predominantly
by the losers. One reason is that analytic philosophy is a somewhat anti-
historical tradition, especially where it most resembles a science, in aspiration
or achievement. For there it tends to be oriented towards the future rather
than the past, in the manner of a science – hardly surprising when progress is
expected. Those who do not like history cannot complain when their history is
written by people who are not like them. A second reason is that recent analytic
philosophy seems to subvert the global narratives it might otherwise be tempting
to tell about the history of the subject – most notably, in the resurgence of realist
metaphysics, often unashamedly concerned with things in themselves. For those
sympathetic to Kant or Wittgenstein or Dewey, it must be tempting to see much
recent analytic philosophy as an insignificant anomaly, a passing throwback, in
the long march of philosophy.
A case in point is Richard Rorty, who was admirably willing to step back,
identify bold patterns in the then-recent history of analytic philosophy, and list
his heroes – Kant, Hegel, Wittgenstein, Dewey, Heidegger, Sellars, Brandom, ...;
no wonder his racy, deliberately provocative stories have been so widely read. It
is striking that the very large number of names of contemporary philosophers
– villains as well as heroes – in the index to Philosophy and the Mirror of
Nature (Rorty 1979) does not contain that of David Lewis, who had already
published two highly much-discussed books and many articles, and been Rorty’s
colleague at Princeton since 1970. Rorty’s radar had missed a serious threat, the
central figure in analytic philosophy for the coming decades. Rorty was out of
sympathy with most new wave philosophy of language, and the metaphysics
that increasingly accompanied it, because its referential approach to semantics
came too close for his comfort to making language a mirror of the world. For
the future, he put his money instead on the inferential approach, particularly
in the neo-pragmatist form offered by Robert Brandom (1994), focussed on
the commitments and entitlements of speakers to make moves in the language
game. Brandom himself has his own grand narrative of the history of philosophy,
in which – tongue partly but not wholly in cheek – he presents himself as the
natural successor to Kant and Hegel (Brandom 2009). But his inferentialism
has remained at an even more programmatic stage than Dummett’s, lacking
an equivalent of Dummett’s connection with technical developments in proof
theory by Dag Prawitz and others. As a result, inferentialism has been far less
fruitful than referentialism for linguistics. In that crude sense, referentialism
beats inferentialism by pragmatic standards.
Of course, we cannot expect a history of recent philosophy to remain
neutral about the future. Even the driest chronicle of who published what when
has implicit standards of historical significance in selecting whom and what
to include. Good historical narratives discern patterns in their material more
explicitly and reflectively. This paper – which manifestly does not aspire to the
depth or rigour of serious history – has indicated a few of the messy complexities
which any history of recent analytic philosophy must try to order. Nevertheless,
it does at least gesture towards some larger patterns to be made explicit and
reflected on. A chronicle is not enough.
The power of fashion in philosophy already ensures that its history will
exhibit some patterns, if only of the mob rushing here and there. Some of
those fashions look foolish in retrospect; most of them did at the time to
non-sympathizers. But fashion is powerful in all academic disciplines, even in
mathematics – for instance, concerning which branches of the subject or styles
of work carry most prestige. Nor is that merely an inevitable defect in any
collective human enterprise. Academic fashions arise because people trained in a
discipline have some respect for the judgment of others trained in the discipline
as to what is good or fruitful work, worth imitating or following up. When things
go well, that mechanism enables the community to concentrate its energies
quickly where progress is being and will be made, to avoid wasted effort, and to
raise collective standards. It is a way of learning from others. The word ‘fashion’
is most appropriate when the level of deference to majority opinion becomes
too high, stifling diversity and independence of mind, making it harder in the
long run for the community to back up out of a wrong turning, since it loses its
sense of the alternatives. But the rule of fashion is only an exaggerated form of
something no community can do without. Even the time and energy spent on
bad ideas and misconceived programmes has its value, since the effect of the
investment is that their limitations are properly explored and tested, so lessons
are properly learnt. The history of academic fashions is the history of how things
once looked to highly intelligent and knowledgeable people.
The changes in philosophy discussed in this paper occurred in a period whose
political, social, and cultural history is already being written. Its philosophical
history needs to be properly written too, by historians of philosophy with at least
enough sympathy for them to understand why what so many philosophers did
seemed a good idea at the time. There are encouraging signs that such histories
are just starting to be written. I look forward to reading them.
References
Austin, J. L. 1956–7: ‘A plea for excuses’, Proceedings of the Aristotelian Society, 57,
pp. 1–30.
Ayer, A. J. 1936: Language, Truth and Logic. London: Victor Gollancz. Page
references to the 2nd edition, 1946.
Baker, Gordon, and Hacker, Peter 1984: Frege: Logical Excavations. Oxford: Blackwell.
Bradley, F. H. 1893: Appearance and Reality. London: Swan Sonnenschein.
Brandom, Robert 1994: Making it Explicit: Reasoning, Representing, and
Discursive Commitment. Cambridge, Mass.: Harvard University Press.
Brandom, Robert 2009: ‘Animating ideas of idealism: a semantic sonata in Kant
and Hegel’, in his Reason in Philosophy. Cambridge, Mass.: Harvard University
Press.
Carnap, Rudolf 1947: Meaning and Necessity: A Study in Semantics and Modal
Logic. Chicago: University of Chicago Press.
Chomsky, Noam 1957: Syntactic Structures. The Hague: Mouton.
Chomsky, Noam 1959: ‘Review of B. F. Skinner’s Verbal Behavior’, Language, 35,
pp. 26–58.
Davidson, Donald 1967a: ‘Truth and meaning’, Synthese, 17, pp. 304–323.
Davidson, Donald 1967b: ‘The logical form of action sentences’, in N. Rescher (ed.),
The Logic of Decision and Action. Pittsburgh: University of Pittsburgh Press.
Davidson, Donald 1973–74: ‘On the very idea of a conceptual scheme’, Proceedings
and Addresses of the American Philosophical Association, 47, pp. 5–20.
Davidson, Donald 1977: ‘The method of truth in metaphysics’, in P. A. French,
T. E Uehling, Jr., and H. K. Wettstein (eds.), Midwest Studies in Philosophy, 2:
Studies in the Philosophy of Language. Morris: University of Minnesota Press.
Dennett, Daniel 1981: Brainstorms: Philosophical Essays on Mind and Psychology.
Brighton: Harvester.
Dummett, Michael 1958–9: ‘Truth’, Proceedings of the Aristotelian Society, 59, pp.
141–162.
Dummett, Michael 1973: Frege: Philosophy of Language. London: Duckworth.
Dummett, Michael 1975: ‘Wang’s paradox’, Synthese 30, pp. 301–324.
Dummett, Michael 1978: ‘Can analytical philosophy be systematic, and ought it
to be?’, in his Truth and Other Enigmas. London: Duckworth.
Dummett, Michael 1984: ‘An unsuccessful dig’, in C. Wright (ed.), Frege: Tradition
and Influence. Oxford: Blackwell.
Dummett, Michael 1991: The Logical Basis of Metaphysics. London: Duckworth.
Evans, Gareth 1982: The Varieties of Reference, ed. J. McDowell. Oxford:
Clarendon Press.
Fine, Kit 1975: ‘Vagueness, truth and logic’, Synthese 30, pp. 265–300.
Fodor, Jerry 1975: The Language of Thought. New York: Crowell.
Grice, H. P. 1961: ‘The causal theory of perception’, Aristotelian Society, sup. 35,
pp. 121–52.
Grice, H. P. 1975: ‘Logic and conversation’, in D. Davidson and G. Harman (eds.),
The Logic of Grammar. Encino: Dickenson.
Grice, H. P., and Strawson, P. F. 1956: ‘In defence of a dogma’, Philosophical
Review, 65, pp. 141–158.
Hawthorne, John 2004: Knowledge and lotteries. Oxford: Oxford University Press.
Kripke, Saul 1963: ‘Semantical considerations on modal logic’, Acta Philosophica
Fennica, 16, pp. 83–94.
Kripke, Saul 1972: ‘Naming and necessity’, in D. Davidson and G. Harman (eds.),
Semantics of Natural Language. Dordrecht: Reidel. Revised version with
editions published as his Naming and Necessity. Oxford: Blackwell, 1980.
Kripke, Saul 1976: ‘Is there a problem about substitutional quantification?’, in G. Evans
and J. McDowell (eds.), Truth and Meaning. Oxford: Oxford University Press.
Kripke, Saul 1982: Wittgenstein on Rules and Private Language: An Elementary
Exposition. Oxford: Blackwell.
Kripke, Saul 2013: Reference and Existence: The John Locke Lectures. Oxford:
Oxford University Press.
Lewis, David 1968: ‘Counterpart theory and quantified modal logic’, Journal of
Philosophy, 65, pp. 113–126.
Lewis, David 1969: Convention: A Philosophical Study. Oxford: Blackwell.
Lewis, David 1974: ‘Radical interpretation’, Synthese, 23, pp. 331–344.
Lewis, David 1975: ‘Languages and language’, in K. Gunderson (ed.), Language,
Mind, and Knowledge. Minneapolis: University of Minnesota Press.
Lewis, David 1983a: Philosophical Papers, Volume I. Oxford: Oxford University Press.
Lewis, David 1983b: ‘New work for a theory of universals’, Australasian Journal
of Philosophy, 61, pp. 343–377.
Lewis, David 1986: On the Plurality of Worlds. Oxford: Blackwell.
Montague, Richard 1970: ‘English as a formal language’, in B. Visentini et al.
(eds.), Linguaggi nella Società e nella Tecnica, Milan: Edizioni di Comunità.
Montague, Richard 1973: ‘The proper treatment of quantification in ordinary
English’, in J. Hintikka, J. Moravcsik, and P. Suppes (eds.), Approaches to
Natural Language: Proceedings of the 1970 Stanford Workshop on Grammar
and Semantics. Dordrecht: Reidel.
Nagel, Thomas 1974: ‘What is it like to be a bat?’, Philosophical Review, 83, pp.
435–450.
Peacocke, Christopher 1992: A Study of Concepts. Cambridge, Mass.: MIT Press.
Prior, A. N. 1957: Time and Modality. Oxford: Clarendon Press.
Putnam, Hilary 1978: Meaning and the Moral Sciences. London: Routledge &
Kegan Paul.
Putnam, Hilary 1994: ‘Sense, nonsense, and the senses: an inquiry into the
powers of the human mind’, Journal of Philosophy, 91, pp. 445–517.
Quine, Willard van Orman 1951: ‘Two dogmas of empiricism’, Philosophical

Review, 60, pp. 20–43.
Quine, Willard van Orman 1960: Word and Object. Cambridge, Mass.: M.I.T. Press.
Quine, Willard van Orman 1969: Foreword to Lewis 1969.
Rorty, Richard 1979: Philosophy and the Mirror of Nature. Princeton: Princeton
University Press.
Salmon, Nathan 1982: Reference and Essence. Oxford: Blackwell.
Searle, John 1969: Speech Acts: An Essay in the Philosophy of Language.
Cambridge: Cambridge University Press.
Stanley, Jason 2005: Knowledge and Practical Interests. Oxford: Clarendon Press.
Strawson, P. F. 1950: ‘On referring’, Mind, 59, pp. 320–324.
Strawson, P. F. 1952: Introduction to Logical Theory. London: Methuen.
Strawson, P. F. 1959: Individuals: An Essay in Descriptive Metaphysics. London:
Methuen.
Strawson, P. F. 1971: ‘Meaning and truth’, in his Logico-Linguistic Papers. London:
Methuen.
Strawson, P. F. 1977: ‘Analitička filozofija’, Theoria (Belgrade), 12, pp. 61–71.
Strawson, P. F. 1992: Analysis and Metaphysics: An Introduction to Philosophy.
Oxford: Oxford University Press.
Strawson, P. F. 1998: ‘Intellectual autobiography’, in L. Hahn (ed.), The Philosophy
of P.F. Strawson. Chicago: Open Court.
Wiggins, David 1976: ‘Truth, invention, and the meaning of life’, Proceedings of
the British Academy, 62, pp. 331–378.
Williamson, Timothy 1988: ‘First-order logics for comparative similarity’, Notre
Dame Journal for Formal Logic, 29, pp. 457–481.
Williamson, Timothy 1984: ‘The infinite commitment of finite minds’, Canadian
Journal of Philosophy, 14, pp. 235–255.
Williamson, Timothy 1994: Vagueness. London: Routledge.
Williamson, Timothy 2013: Modal Logic as Metaphysics. Oxford: Oxford
University Press.
Williamson, Timothy 2014: ‘Replies to Bricker, Divers, and Sullivan on Modal Logic
as Metaphysics’, Philosophy and Phenomenological Research, 88, pp. 744–764.
Wilson, John Cook 1926: Statement and Inference. Oxford: Clarendon Press.
Wright, Crispin 1975: ‘On the coherence of vague predicates’, Synthese, 30, pp.
325–365.
DELIBERATIVE DEMOCRACY
Jürg Steiner Original Scientific Paper

University of North Carolina at Chapel Hill and UDK 316.454.5:141.7
University of Bern
Maria Clara Jaramillo
University of Cali, Colombia
Simona Mameli
University of Bern
THE DYNAMICS OF DELIBERATION
Abstract: Existing instruments to measure the quality of deliberation are too static,
focusing too much on the analysis of the individual speech acts. We present an
instrument to identify Deliberative Transformative Moments (DTM) lifting the
level of deliberation from a low to high or vice versa. To use this instrument, one
has to look at the group dynamics of the entire discussion. Empirical basis are
discussions among Colombian ex-combatants from both the extreme left and the
extreme right. We investigate to what extent personal stories have either a positive
or a negative impact on deliberative transformative moments. A corresponding
typology of personal stories is developed.
This paper is a continuation of previous research on the deliberative model

of democracy.1 Good scholarship should ideally be creative destruction of one’s
previous research.2 This is precisely what we want to do in our current research,
although not in an extreme form. In our previous research, we developed the
method of Discourse Quality Index (DQI) to measure the deliberative quality
of the speech acts of discussions in parliaments and also among ordinary
citizens. The values for the individual speech acts were aggregated for the
individual participants and also for the discussion groups at large. Theoretically,
1 Jürg Steiner, André Bächtiger, Markus Spörndli, and Marco R. Steenbergen, Deliberative
Politics in Action. Analysing Parliamentary Discourse, Cambridge University Press, 2005.
Jürg Steiner, The Foundations of Deliberative Democracy. Empirical Research and Normative
Implications, Cambridge University Press, 2012.
2 Analogous to what Joseph Schumpeter postulated for the economy; Joseph Schumpeter,
Capitalism, Socialism and Democracy, third edition, New York: Harper and Row, 1950.
we could explain variation in these aggregated values in their antecedents and

consequences. The measurement instrument of the DQI has now been widely
used all over the world.3 We consider it still as useful instrument for many
purposes, but acknowledge that it is too static to get at the dynamics of group
discussions. We have now created the concept of Deliberative Transformative
Moment (DTM) to identify situations when the group dynamics lead to an
upswing or a downswing in the level of deliberative quality. The purpose of this
paper is to introduce this new concept into the deliberative literature.
As academics, we know intuitively from our everyday experience of sitting
in all kind of meetings that all of a sudden the tone of a meeting may change,
getting either more deliberative or less so. Thus, at the intuitive level, we are
familiar with the concept of deliberative transformative moments. In this paper,
we attempt to get more systematically at such situations. On the one hand, we try
to identify situations where the discussion drags on at a low level of deliberation,
and then something occurs to transform it to a high level. On the other hand, we
investigate situations where the discussion flows at a high level of deliberation,
and then something occurs to transform it to a low level. Our basic theoretical
question is what this something can be. First, we want to show why the DQI
is not sufficient to get at these dynamics.4 Then, we will show how we identify
Deliberative Transformative Moments (DTM).
The DQI measures the various deliberative dimensions. The units of
analysis are the individual speech acts. Each speech act is coded according to
given categories for each dimension. One dimension is how well arguments are
justified. The coding categories refer to how well reasons and conclusions are
linked. Personal stories also count as good justifications as long as they are linked
to the issue under discussion. A second dimension refers to the respect that is
paid to other actors and the arguments they present. A third dimension asks to
what extent arguments are justified in terms of the public good. Self-interest are
compatible with good deliberation, especially if they come from underprivileged
persons and groups. A fourth dimension has to do with the outcome of a group
discussion. From a deliberative perspective, consensus is a good outcome, but
it may be good enough if the actors acknowledge that the other side also has
valid arguments. A fifth dimension asks whether all actors are free to speak
up or whether they are constrained, especially by unwanted interruptions or
other intimidations. The last dimension deals with the question whether actors
actually mean what they say. This question of truthfulness is most difficult to get
an empirical handle on. Crude lies are usually easy to detect, but otherwise the
DQI limits itself to the question whether actors perceive each other as truthful.
Coding all speech acts of a discussion according to these dimensions
gives a good overview of the deliberative quality of the discussion. Initially, we
developed the DQI for parliamentary debates in Germany, Switzerland, the UK
3 For an overview of the place of the DQI in contemporary deliberative research see Simon
Beste, “Contemporary Trends of Deliberative Research,” Journal of Public Deliberation 9
(2013), issue 2.
4 The DQI is presented in the appendix of both books cited in footnote 1.
Jürg Steiner, Maria Clara Jaramillo, Simona Mameli: Te dynamics of deliberation 41
and the US, both for plenary sessions and committee meetings.5 Speech acts in
parliamentary debates have a high formality with the chair giving the floor to
one actor after another. Thus, parliamentary speech acts have usually a certain
length, which allows scholar to use the DQI to get at the dynamic aspects of
a debate. André Bächtiger et al. have done this in a fruitful way for a Swiss
parliamentary committee that discussed over eight sessions a language bill.6 They
found, for example, that at first many actors told personal stories and that this
storytelling greatly diminished over time. A further finding was that references
to the common good and rational justifications also decreased over time.
The research situation is very different when we investigate informal group
discussions of ordinary citizens. This was the case for our study of discussions
between ex-combatants of the extreme left and the extreme right in Colombia
and of Serbs and Bosnjaks in Srebrenica in Bosnia-Herzegovina.7 At the
beginning of the discussions, moderators indicated that the topic for discussion
was to find ways for peace and then let the discussion go wherever it went.
The consequence was an often quick interactive pattern with many shortcuts,
a pattern very different from formalized parliamentary debates. Sometimes, it
happened that a participant uttered only a single word. According to the DQI,
the discourse quality of such a speech act would be extremely low. From the
perspective of Deliberative Transformative Moments (DTM), however, it would
all depend on the context in which such a word is uttered. Let us discuss an
example, where the utterance of a single word led the discussion to continue
at a high level of deliberation. The example stems from a group of Colombian
ex-combatants. Arturo, an ex-guerrilla, uttered the single word rehabilitation.
What was the context in which this word was uttered? Before Arturo made
this extremely short intervention, the discussion flowed at a high level of
deliberation. The group addressed the issue whether the constitution should be
amended allowing the death penalty for rapists. In a very interactive way, the
group discussed this issue, addressing also alternatives like castration and life
in prison. Arguments were justified, and participants showed respect for the
arguments of others. Immediately before Arturo uttered the word rehabilitation,
Bernardo, an ex-paramilitary, made the following statement:
Family is the nucleus of society. I see Colombia as a big family, and if I make
a mistake and my brother goes to my father and tells him to beat me up, then we
are not doing anything good. What we have to do is to provide the mechanisms
and the means for that person to be able to realize the bad things he is doing and
completely change his behaviour.
Bernard brings the alternative into the discussion that rapists should not
be punished but should be helped to change their behaviour. To support his
5 Steiner et al., Deliberative Politics in Action.
6 André Bächtiger, Shawn Rosenberg, Seraina Pedrini, Mirjam Ryser, and Marco R.
Steenbergen, “Discourse Quality Index 2: An Updated Measurement Instrument for
Deliberative Processes,” Paper Presented at the 5th ECPR General Conference, Potsdam,
September 10–12, 2009.
7 Steiner, The Foundations of Deliberative Democracy, pp. 15–21.
argument, he claims that in a family it does not help when the father beats up
a son, and he compares Colombia to a family. With this statement, Bernard
keeps up the flow of the discussion, staying on topic and moving the discussion
forward in adding another alternative, for which he gives a justification. Thus,
the discussion stays at a high level of deliberation. What is Arturo attempting
to accomplish when at this point he utters the word rehabilitation? Given the
context, our interpretation is that Arturo gives Bernard a helping hand in
telling the group that what Bernard suggests goes under the technical term
of rehabilitation. With this helping hand, Arturo does not disrupt the flow of
high deliberation. He clarifies for the group what Bernard suggests and, in this
way, helps the discussion to continue on a more solid basis of knowledge. The
discussion indeed continues to flow at a high level of deliberation. Coding the
one-word speech act of Arturo with the DQI would give the impression that
the level of deliberation had sharply dropped. According to our interpretation,
however, uttering the word rehabilitation did not at all disrupt the flow of high
deliberation but fitted well into its flow. With our approach, we look at how the
individual speech acts are linked with each other.
The challenge in our research is to identify situations, where the discussion
is transformed to a high level of deliberation or drops to a low level. How do
we distinguish between the two levels? We take a qualitative approach, which
has much to do with linguistics and social psychology. We look at what actors
may wish to accomplish when they utter words in a particular context, and then
we attempt to determine whether a transformative moment occurs. When a
discussion drags on at a low level of deliberation, we look for mechanisms by
which the discussion finds its way to a high level of deliberation. Thereby, it is
not only important what speakers intend to say but also how listeners interpret
what the speakers say. When, on the other hand, the discussion flows at a high
level of deliberation, we look for mechanisms by which this flow is disrupted, and
the discussion is transformed to a low level of deliberation. Given this research
agenda, we have for each speech act four coding categories:
1) The speech act stays at a high level of deliberation.
2) The speech act transforms the level of deliberation from high to low
3) The speech act stays at a low level of deliberation
4) The speech act transforms the level of deliberation from low to high
The coding situation is not fundamentally different as for the DQI. In both
cases, one makes a coding judgment for each speech act. For the DQI, one codes
each speech act per se, for example, to what extent foul language is used or how
well arguments are justified.8 For Deliberative Transformative Moments (DTM),
we look at the entire context to determine which of the four coding categories
applies. For both research projects, one has to make judgments that have to some
extent a subjective nature. Therefore, it is not a question of quantitative versus
qualitative research; in both cases one has to make qualitative judgments. For the
DQI, one has much more coding data, since one codes several dimensions of the
8 Steiner et al, Deliberative Politics in Action. Steiner, The Foundations of Deliberative Democracy.
discourse quality. Given the wealth of coding data, the usual publishing practice
is that only the coding decisions are revealed but not their justification, or at
most with some sample examples. The focus tends to be on complex statistical
analyses showing the antecedents and consequences of variation in the DQI.
For the current project, we are not interested in statistical analyses. Our
research strategy is rather to give a full description of how we arrived at our coding
decisions. We tell in an in-depth way the story of each group discussion with
its ups and downs. The reader is invited to follow closely how we interpret the
dynamic that goes on in a discussion. Since our interpretations are fully revealed,
the reader has the basis to take a different view on what is going on in a group.
With this approach we are close to Ron Lubensky, who analysed the discussions
of the Australian Citizens’ Parliament (ACP).9 The title of his paper already
indicates in what direction he wants to go with his analysis: „Listening Carefully
to the Citizens’ Parliament: A Narrative Account.” He wishes „to open a window
to the story of the ACP’s participants.”10 Lubensky does not claim that he has „a
master story from which all interpretations of the ACP should follow, nor (is he)
claiming that the story line presented here is the only one.” His main point is „that
a reflective, storied approach to analysing the events, based on narrative methods
of discourse analysis, provides useful insight into the process and capacities of
participants.” This is pretty much what we have in mind in our current research.
In practical terms, we proceed in our coding in such a way that we try to put
ourselves in the context in which each actor actually speaks. This means that we
consider only the speech acts that are already uttered and not those that follow.
Times and again, we went back to what was said before, checking the tapes and
the transcripts making sure that we had a good feeling for the context in which
an actor intervened in this discussion. Only when we had made a coding decision
did we proceed to the next speech act. In this way, we try to follow the narrative
of the discussion quasi life, which means as it is experienced by the participants
themselves, who obviously do not know what will happen after they speak.
That one should not look at individual speech acts in isolation but in how
they relate with what was said before is also emphasized by conversation theory,
which is prominent among anthropologists. We became only recently aware
of this theory and find it comforting to get support for our approach from
outside the deliberative literature. In a review paper, Charles Goodwin and John
Heritage put the core of conversation theory in these terms: „conversational
participants will inevitably display some analysis of one another’s actions. Within
this framework of reciprocal conduct, action and interpretation are inextricably
intertwined ... in the real world of interaction sentences are never treated as
isolated, self-contained artefacts.”11 Conversational captures well what is also our
intention to analyse discussions as an interactive process.
9 Ron Lubensky, “Listening Carefully To The Citizens’ Parliament: A Narrative Account,” in
Carson L, Gastil J., Hartz-Karp J, Lubensky R., (eds.), The Australian Citizens’ Parliament
and the Future of Deliberative Democracy, University Park. PA: Pennsylvania State University
Press, 2013.
10 This and the following quotes Lubensky, “Listening Carefully,” p. 66.
11 Charles Goodwin and John Heritage, “Conversation Analysis”, Annual Review of Anthropology,
Vol. 19 (1990), pp. 287–8.
Although we do not look for authoritative stories, we are still interested

how much reliability we get in our own coding judgments. We checked it for a
randomly selected group of Colombian ex-combatants, where 107 speech acts
were uttered. Each speech act was coded independently by Maria Clara Jaramillo
and Jürg Steiner. The former is from Colombia, the latter from Switzerland. It
is important that there is someone who is familiar with the local Colombian
culture; but it is equally important to have someone who can look into the
Colombian culture from a different cultural background. Given these different
perspectives, it is comforting that reliability was high with an agreement between
the two coders in 98 of the 107 speech acts. We acknowledge, of course, that the
two of us have worked for many years together, so that it has to be expected that
we tell the story of this discussion group in a similar way. Despite this caveat, the
high reliability may still increase the trustworthiness of our results, at least for
ourselves, and perhaps also for the readers.
How do we put our four coding categories in operational terms? Basically, for
the discussion to flow at a high level of deliberation it must be somehow linked
to the peace process, the topic assigned to the groups both in Colombia and
Bosnia-Herzegovina. By contrast, the discussion is at a low level of deliberation
if it drags on off topic without any clear direction. Deliberative Transformative
Moments (DTM’s) occur when the discussion changes from one level to the
other. Specifically, we use the following criteria for our coding decisions:
Code 1: The speech act stays at a high level of deliberation
The coding is easiest if a speech acts fulfils all the criteria of the DQI
mentioned above, which means that the speaker has not interrupted
other speakers, justifies arguments in a rational way or with relevant
stories, refers to the public good, respects the arguments of others and
is willing to yield to the force of the better argument. The discussion,
however, can still continue at a high level of deliberation if speakers
do not fulfil all these criteria as long as they stay in an interactive way
on topic. If a speaker, for example, simply supports the argument of
a previous speaker without adding anything more, the discussion
continues to flow at a high level of deliberation. An extreme example is
the one word intervention of Arturo mentioned above. For our coding
decisions the concept of topic has particular importance, by which
we mean a subject matter that has a certain internal consistency. An
example of a topic that we encountered in the discussions of Colombian
ex-combatants is poverty in the country. As long as a speech act stays
within this topic, even if the speech act is very brief and not very
elaborate, the level of deliberation remains high. Our criterion is whether
the discussion continues to flow in an interactive way on a particular
topic with the actors listening to each other with respect. Deliberation
also stays high if an actor introduces another topic and presents it in
such a way that the other actors can understand it, especially in giving
reasons why the topic is linked with the peace process. An actor may,
for example, turn the discussion from poverty to corruption, a case that
actually occurred in one of the groups of Colombian ex-combatants.
If the issue of corruption is sufficiently linked to the peace process the

discussion continues at a high level of deliberation.
Code 2: The speech act transforms the level of deliberation from
high to low
We use this second coding category when the flow of the discussion
is disrupted. The topic debated so far is no longer pursued, and no
new topic related to the peace process is put on the agenda. Topics
and stories may be mentioned that have nothing to do with the peace
process and are therefore completely off topic. It is also possible that the
speech act is so incoherent and confusing that it does not make sense.
Under these various circumstances, there is no open space where other
actors could easily continue the discussion in a directed way.
Code 3: The speech act stays at a low level of deliberation
We use the third coding category for speech acts that do not manage to
give to the discussion again a direction linked with the peace process.
The speaker is unable or unwilling to put on the agenda a topic relevant
for the peace process. Instead, the speaker brings up topics or stories that
are off topic, or the speech act is altogether incoherent and confusing.
The key criterion for this third coding category is that the speech does
not open windows for the group to talk about the peace process.
Code 4: The speech act transforms the level of deliberation from low
to high
Speech acts coded according to this fourth category are successful in
formulating a new topic relevant for the peace process. Success means
that good arguments are presented why the topic should be discussed
in connection with the peace process. In this way, the speech acts opens
new space for the discussion to continue in a directed way.
Having established what we mean by Deliberative Transformative Moments
(DTM), the further research task will be to investigate what happens after such
moments. If they go upwards, we want to see how efficient they are, by which we
mean how long afterwards the discussion continues to flow at a high deliberative
level. For downward Deliberative Transformative Moments (DTM), we will
check how detrimental they are, by which we mean how long afterwards the
discussion drags on at a low level of deliberation. A further research task will be
to explain the strong variation in the frequency of Deliberative Transformative
Moments (DTM) among the various groups that we study.
At the beginning of the paper, we claim that the concept of deliberative
transformative moment is new in the deliberative literature. We acknowledge,
however, that Simon Niemeyer comes close to the concept, when in his PhD
dissertation he writes about „turning points” in deliberation.12 We also note
that Lyn Carson reports that a participant in the discussions of the Australian
12 Simon Niemeyer, Deliberative Monetary Valuation as a Political Economic Methodology:

Exploring the Prospect for Value Pluralism with a Case Study on Australian Climate Change
Policy, PhD dissertation, Australian National University, 2011.
Citizens’ Parliament actually talks about a „transformative” incident, when

something unusually had happened changing the tone of the deliberations.13
Thus, it is not quite unknown in der deliberative literature that something
like deliberative transformative moments exist, although the concept has not
yet been fully developed. Outside the deliberative literature, the concept of
catharsis has some similarities with our concept of transformative moments. It
was initially presented by Aristotle in his response to Plato’s critics on drama.
According to Plato, drama should be closely controlled or eliminated, as it fosters
human passions. Aristotle, on the contrary, argued that „dramatic catharsis was
necessary, that it purged the audience of pity and terror14„ In fact, in his Poetics,
Aristotle argues that „drama tends to purify the spectators by artistically exciting
certain emotions, which act as a kind of homeopathic relief from their own
selfish passions.15„ To be relieved from selfish passions fits well the situations
when a discussion is transformed to a higher level of deliberation.
The concept of catharsis was also used in the early development of
psychiatry. At the group level, Jacob L. Moreno explains how it was applied in
the work on psychodrama that began in Vienna in the 1920’s. „This change has
been exemplified by the movement away from the written (conserved) drama
and toward the spontaneous (psycho) drama with the emphasis shifting from
the spectators to the actors.”16 Moreno’s work is based on group dynamics and
is indeed closely related to the idea of transformative moment. In fact, through
a process of catharsis a change is produced in the actors in the psychodrama
very similar to our concept of transformative moment. At the individual level,
Thomas Scheff and Don D. Bushnell describe how Sigmund Freud in his studies
of hysteria presented catharsis as a „quick, cheap and effective cure for hysterical
neurosis.”17 Scheff and Bushnell propose a theory of catharsis that involves three
interacting systems, one biological, one psychological and one social. From
the perspective of our own research, it is interesting to see how the concept of
catharsis has a dynamic aspect similar to our concept of transformative moment.
Turning to contemporary research, the concept of transformation is
particularly prominent in conflict and peace studies. John Paul Lederach uses
the term conflict transformation, by which he means that a conflict is „a potential
catalyst for growth.”18 Johan Galtung also speaks of conflict transformation; for
him „to transform a conflict is to transplant it into a new reality”19 Conflict
transformation is present when, accepting that conflict is both a source of
13 Lyn Carson, “Investigation of (and Introspection on) Organizer Bias,” in Carson L, Gastil
J., Hartz-Karp J, Lubensky R., (eds.), The Australian Citizens’ Parliament and the Future of
Deliberative Democracy, University Park. PA: Pennsylvania State University Press, 2013.
14 Thomas J. Scheff and Don D. Bushnell, “A Theory of Catharsis,” Journal of Research in
Personality 18 (1984), 238
15 Jacob L. Moreno, “Mental Catharsis and the Psychodrama,” Sociometry 3, 3 (1940), 209.
16 Jacob L. Moreno, “Mental Catharsis and the Psychodrama,” Sociometry 3, 3 (1940), 209.
17 Thomas J. Scheff and Don D. Bushnell, “A Theory of Catharsis,” Journal of Research in
Personality 18 (1984), 238.
18 John P. Lederach. The Little Book of Conflict Transformation. Good Books (2003). p.15
19 Johan Galtung. Conflict Transformation By Peaceful Means: A Trainer’s Manual. United
Nations (2000). p. 4
creation and a source of destruction, we decide to act in such a way that the
creative aspects take control. The task of transforming a conflict requires
„finding positive goals for all parties, imaginative ways of combining them, and
all of this without violence.”20
In education, Jack Mezirow understands transformative learning as „the
process of effecting change in a frame of reference.”21 Frames of reference, in
turn, are the structures of assumptions through which we understand our
experiences.”22 Finally, the United Nations Educational, Scientific and Cultural
Organization – UNESCO – uses the concept of social transformation to describe
a critical stance towards traditional notions of development. It „considers social
transformation research as a field of research that can lead to positive steps for
social and political action to protect local and national communities against
negative consequences of global change.”23
These references to the scholarly literature at large indicate that the concept
of transformation has a prominent place in many fields. Thereby, it is important
to note that transformation is not simply a synonym for change but denotes a
particular sort of change, namely an abrupt change in the sense of a catharsis. It
is such transformations that form the topic of our new research. To present such
research will be for another place. In this paper we only wished to introduce
Deliberative Transformative Moments (DTM) as a new concept, of which we
hope that it will be fruitful to get at the dynamic aspects of deliberation.
20 Johan Galtung. Conflict Transformation By Peaceful Means: A Trainer’s Manual. United

Nations (2000). p. 3
21 Jack Mezirow. “Transformative Learning: Theory to Practice.” New Directions in Adult and
Continuing Education 74 (1997) p.5
22 Jack Mezirow. “Transformative Learning: Theory to Practice.” New Directions in Adult and
Continuing Education 74 (1997) p.5
23 UNESCO. United Nations Educational, Scientific and Cultural Organization. Social
Transformation. [online] Retrieved from: http://www.unesco.org/new/en/social-and-human-
sciences/themes/international-migration/glossary/social-transformation/
Milorad Stupar Original Scientific Paper
University of Belgrade UDK 321.7:141.7
321.01:17.02
DEMOCRACY AND TRUTH:

A CLUSTER OF PROBLEMS  AN ATTEMPT
TO SORT THEM OUT1
Abstract: This paper attempts to show that an idea of truth and a normative
liberal democratic theory can go hand in hand. First, it is argued that truth has
something to do with politics. Second, three possible aspects of the relationship
between democratic government and truth are discussed. The discussion begins
with a presentation of arguments for the view that the process of establishing
scientific truth stricto sensu is not generally a matter of wide public or democratic
debate, and thus is not incompatible with a democratic order. The discussion then
moves on to consider practical social problems; it endswith the claim that unlike
in „pure” scientific inquiry, democratic procedure is directly relevant and perhaps
crucial to solving practical collective problems. The final part of the paper concerns
„the market of ideas” and the possibility of its failure. It is clear that market of ideas
is not sufficient for producing autonomous and critical citizens who have a „strong
desire not to be fooled” (Bernard Williams). Quite the contrary, the marketplace
of ideas may be exposed to agencies that deliberately distort and conceal the
truth. Governments are prone to commit such illegitimate actions because of the
very nature of their ways of working. This may be the most difficult problem for a
democratic institutional framework.
Keywords: truth, politics, democracy
Introduction
Perhaps we should begin our discussion with the alleged dangers of truth
for democracy. For example, many important political philosophers (Hannah
Arendt, Karl Popper, and John Rawls, to name but a few) worried about the
potential authoritarian and totalitarian implications of the concept of truth in
politics. On this view, truth has a coercive nature that is unfriendly to the spirit
of tolerance necessary for democratic political life that acknowledges value-
pluralism in a society. In a similar vein but much earlier, John Locke – in his
Letter Concerning Toleration – expressed the view that the function of law is not
1 Work on this paper was supported by the project “Dynamic Systems in Nature and Society:
Philosophical and Empirical Aspects” (No. 179041), financed by Ministry of Education,
Science and Technological Development of the Republic of Serbia.
to secure the truth of citizens’ opinions but to provide stable and longlasting
government where believers of various kinds can simultaneously practice their
religious commitments.
Suppose that for certain reasons we want to keep the term ‘truth’ in a
theoretical debate about the nature of the political. Offhand, it seems that politics
has something to do with truth. An argument can be developed in the following
way: assuming that life’s being worth living has something to do with the
conditions of the world we inhabit, which may be so because most of us want to
cultivate the world around us by cooperating with others in light of, say, visions
of the good life, it seems that we should be concerned with having adequate
information about the world, that is to say, we cannot avoid truth claims, or, at
least, some probabilistic estimates about the way the world is. In addition, we
should see to it that the structure of our collective endeavour to attain a good life
fits the demands of the task. In this sense we can say that a political structure is
bad or good, and that the outcome of the working of its institutions are successes
or failures. And this may a way to pull truth between Scylla and Charybdis of
the abovementioned liberal worries. Perhaps the answer is to rely on minimal,
commonsensical, and pragmatic conceptions of truth. Instead of using the term
‘truth’ in a „heavy realist” manner, we can start out with the idea of hypothetical
objectivity (or, with an „independent standard of correctness”)2 in order to
make sense of what we do in science and politics. Both scientific inquiry and
politics are understood as problem-solving activities; „standards of correctness”
in their various forms and character pertain to these activities. The question is
how democratic society and requests for truth understood in this „light” sense
interlock. That is to say, we then have to ask how truth comes into play into
politics, and what democratic politics needs to do such that truth comes into
politics in the right way.
When answering this question, there are several aspects of the problem that
need to be clarified. I’ll name them one by one as they come naturally, without
false hope of exhausting the list. Let us first begin by delineating the aspects of
the problem. Three aspects will be discussed here: 1) the relationship between
scientific inquiry and democratic society; 2) solving practical social problems;
and 3) the problem of creating autonomous citizens (the relationship between
the „marketplace of ideas” and democratic political culture).
Independent Objectivity as a Target:

From Scientific Theories to Moral Principles
It seems that the way we identify the aspects of the relationship between
truth and democracy depends on how we see the character of truth claims.
There is a huge variety of issues with which politics is concerned. Issues that
demand political action could be classified in a series of items that starts out
with, say, deciding on where and how to build bridges, and then goes over
2 Recent work by Helene Landemore relies on this idea. See her Democratic Reason: Politics,
Collective Intelligence and the Rule of the Many, Princeton, Princeton University Press, 2013.
Milorad Stupar: Democracy and truth: a cluster of problems – an attempt to sort them out 51
problems such as global warming, putting a drug on the market, the choice of
tax-system, going to war, immigration, and then finishes, say, with deciding
on allowing euthanasia in hospitals, same-sex marriage, capital punishment,
and so on. On the one side of this imagined spectrum there are deep moral,
religious and political disagreements and a diversity of citizens’ opinions on how
to solve them, and, on the other, there are opinions and judgements that are
more about natural facts and other scientific claims.3 The imagined spectrum
can be extended by adding scientific inquiry that addresses issues with almost
no political bearing. For example, we can think of a scientist who works on the
discovery of subatomic particles or on structure of DNA who is motivated solely
by theoretical questions about the structure of matter, or, say, the beginning of
life. If this is so, we can define our spectrum along the line that begins from basic
or „pure” scientific inquiry, which has only epistemic significance, and which
continues with examples of applied science with both epistemic and practical
significance, and ends up with activity that addresses practical social problems
that have deep moral significance.
In light of all this, we can generate two claims: first, that the range and
variety of issues corresponds to the domain with which they are associated;
second, that the method of solving the issues, when done correctly, is closely knit
with the nature of the domain–i.e., it is hard to imagine a non-trivial, general
standard across the board.
There is one more point that needs to be made clear. Even if we suppose
that there is no unqualified and stateable truth across the board, taking the form
of a „substantial something” out there in the world that makes all our judgments
true or false, we have to concede that the status of the judgments about the
issues (good/bad judgements; correct/incorrect) depends on how far they are
from something we might hypothetically call „objectivity”. When we speak of
natural science, this term could refer to scientific facts and laws that are the
subject matter of the entire explanatory framework4 within a certain field; but it
can also refer to the normative claims that themselves yield practical objectivity.
These latter normative claims can be understood as putative „objective” political
truths. For example, if we reinterpret Rawls’ standards in A Theory of Justice in
this „truth-type talk”, we can say that whether a normative claim is correct is
matter of whether it is in accordance with principles acceptable to „reasonable”,
„free”, and „equal” citizens in the „original position”; thus we can say that it is
3 There are also disagreements that have not been mentioned, the nature of which are more a
matter of convention–for example, it seems that one would have to accept a majority verdict
on whether to drive on the left-hand side of the road without further pondering the deep
significance of this verdict. But, for example, the issue of euthanasia cannot be solved in
the same manner: even if there were a very reliable procedure as to how to make correct
decision on this issue, it would still make sense for somebody as a moral being to ponder the
significance of its results. I owe this last point to David Estlund; see his “Beyond Fairness and
Deliberation”, p.187 in Deliberative Democracy ed. James Bohman and William Rehg, MIT
Press, Cambridge, Massachusets 1997.
4 We can think of this as a settled paradigm of “normal science” in Kuhn’s sense of this idea,
given in his Scientific Revolutions.
probably true that some instances of the minimal conception of common good,
such as absence of war and famine, Pareto-efficient collective outcomes, the rule
of law, economic prosperity, etc., would be chosen by individuals in a Rawlsian-
type „original position”, and this procedure would perhaps be the right way to
attain them. In this sense the common good as described above would count as
objective value.5
Thus on the one side of the spectrum there are the truths of natural science,
and on the other side there are putative truths of moral or political principles; I
assume that truths along this spectrum could be correct answers to scientific and
political questions; I also assume that correct answers are independent from the
procedures we use to attain them. This „standard-independency” attitude does
not have to commit one to any sort of metatheoretical realism, whether scientific
or moral. It is not that facts out there in the world make our judgments true or
false; rather, practices themselves require certain proper standards on the basis
of which we can attain successful predictions or correct answers in a particular
domain. The proper function of these practices pertain to the nature of the field
itself. Some practices, such as natural science inquiry, require a more exclusive
group to be in charge, while some others, contrary to this, require a more inclusive
group. The same applies to the institutional setup within which the groups work
when they come to grips with an issue: when deciding on the acceptability of
euthanasia, for example, the institutional setup should allow a wide democratic
debate, whereas discussion, say, on the interpretation of experiments in physics
on elementary particles does not. The question is, how should democratic society
and the search for truth interlock along the abovementioned spectrum?
We shall see that, unlike in the case of one domain of scientific inquiry which
we may, following Philip Kitcher, call „pure science”,6 democratic procedure is
directly relevant and perhaps crucial to solving practical collective problems in
the right way as long as we as liberals believe that on average individuals are the
best judges about what is good for them (this is both a moral and an epistemic
assumption), and as long as we believe that the electorate as a whole – by using
both deliberative and agregative methods – has a privileged status in solving
collective practical problems.
An example can clarify this point: for epistemic reasons it should not be
left to the expert(s) to decide for a society whether euthanasia should be legally
allowed; on the other hand, suggestions as to how a bridge should be built,
ceteris paribus, should rely more on estimates undertaken by capable engineers
who could give figures offhand simply by relying on received knowledge from
technical science and its constraints. This of course does not mean that authorities
should not take into account anything else besides engineers’ estimates (for
example, the religious commitments of the population scattered around the
construction site should come to the fore at some point); nor does it mean that
the general public should not take into account experts’ opinions on the possible
5 I follow here Helen Landemore’s idea of political cognitivism and procedure-independent
standards of correctness. See her book Democratic Reason (Chapter 8, Political Cognitivism:
A Defense).
6 Philip Kitcher, Science, Truth and Democracy, Oxford University Press, Oxford, 2001, p. 87.
effects of practicing euthanasia. The point is that experts should have the final
say on how to build a bridge, whereas a democratic deliberation, followed by
voting on referendum, could be necessary for eventually deciding the issue of
euthanasia. And it should be no surprise that things should be done this way.
The issue of euthanasia is morally heavily loaded and concerns every citizen in
a political community; it requires maximal inclusiveness of the population and
thus becomes a statutory or constitutional question.
Now, one might object to this that disagreements can occur even within
a group of professional experts. One might wish to push further and claim
that democratic deliberation, along with majority voting, is the best and most
appropriate procedure in these cases and thus democracy stays with us even
when dealing with „pure” science and experts working within its research
projects. A sceptic might even push further and say that evidence for scientific
claims is always insufficient, such that our decisions about what to believe are
always crucially affected by our moral, social, political, and religious values. We
can add to this problem a worry concerning the argument that science is never
merely the pursuit of truth pure and simple, but rather the pursuit of those truths
that scientists believe to be significant.7 One cannot isolate science from society,
as some of the „scientific faithful” would have it, rather dogmatically.
But the question is whether conflicts about evidence in science require
wide democratic public debate to be resolved. And the answer seems to be „no”,
because whether or not a wide electoral body finds them agreeable is not what
matters. For example, the question of whether an elusive subatomic particle is
detectable is not a moral issue, and thus it does not require public deliberation
on the part of the entire electoral body. In addition, one more point needs to be
added: from the assumption that ordinary citizens have equal right to exercise
their freedom of ethical, religious, and political choice, one cannot conclude
that everyone has an inherent ability to assess particular truth claims in science.
When satisfying the virtue of truthfulness or accuracy in scientific endeavours
(at least as far as „pure” science is concerned), it seems that moral, political, and
religious judgments should be as far as possible kept away from any evaluation
of evidence for conclusions.
First Aspect: Scientific Truth and Democratic Society

Generally, the very nature of scientific results about the natural world do
not depend on whether we agree with them or not. They are not a matter of
accidental circumstance or of shifting majority opinions of the general public.
They are objective truths in the sense that one will always get the same result
in the same experiment. For example, an experiment with gases shows that
its result doesn’t depend on who carries it out, but on independently existing
physical laws relating to gases which say that pressure is a function of volume.
Whether John or Peter, as experimenters, change the volume of the gas does not
7 For example, Kitcher (2001) argues that ‘All kinds of considerations, including moral, social,
and political ideals, figure in judgments about scientific significance’ (p. 86).
matter: if the temperature and amount of gas remain unchanged within a closed
system, the pressure of a gas tends to decrease as the volume of a gas increases.
It also seems that our natural, common-sense epistemic attitude toward
truth, scientific inquiry, and the objective world supports some form of scientific
realism. Or, if we do not want to use the word ‘realism’, we can say that it in some
sense supports some „correspondence” of the scientific theories. For example,
when navigating, the fact that we end up so often where we want to go is best
explained by the simple truth that our maps do represent the organisation of
the land. Another examples abound: there are numerous successful scientific
predictions. For example, planes stay up, a space module gets to the moon,
e-mail goes through, etc. It is hard to conceptualize these successes unless we are
in some sense right about the objective world around us. Thus in some sense,
the „correspondence” aspect of scientific theories always stays with us even if we
reject naive realism. The question is, what is left for democratic constituency to
debate about science if we suppose that an exclusive professional group has the
right to decide about what counts as objective knowledge?
What may concern democracy regarding objective scientific knowledge is
transparency and the accessibility of accumulated knowledge when circumstances
require that the public should be acquainted with its results. As Philip Kitcher
puts it, when we say that scientific enquiry should be free, democracy may
require that investigators be honest in the presentation of their findings, that
conduct of experiments cannot override certain important moral values, etc.8
Let us turn now to some practical political problems.
Second Aspect: Solving a Practical Social Problem--

Complexity, Uncertainty, Moral Sensitivity, and Fallibility
Compared to the subjects matters of mathematics and natural science, social
reality is a much messier domain. As Helene Landemore puts it: „When we
discuss political opinions we are dealing with the domain where human beings
deal with the risk and uncertainty of human life as a collective problem”.9 This is
roughly equivalent to what Aristotle meant when he said in Nicomachean Ethics
that when we are dealing with the domain of „human affairs”, we are dealing
with things that could be different.10 In other words, the political domain
is by definition an area that eludes apodictic judgements, where expertise is
difficult to identify in advance. Going back to the previous example with John
and Peter, we can see that in political matters things are not determined in the
way they are determined in physics. Suppose it is true that progressive taxation
increases social welfare among blue-collar workers. From this it does not follow
that John and Peter, who are respected „authoritarian epistocrats” in country
C, will necessarily manage to increase social welfare in C by implementing
8 See Kitcher (2001), p. 3.
9 See Lanemore (2013), p. 13.
10 Compare Aristotle’s discussion of phronesis in Nicomachean Ethics 1140a24–1140b12.
measures of progressive taxation. We wouldn’t be very surprised if they failed as

politicians or economists, so much we would have been surprised if they failed
as experimenters in our previous example with gases in physics.
Politics is different from scientific inquiry because scientific inquiry requires
conceptual and methodological rigor, with skills and knowledge usually acquired
by special training in a system of very regulated institutions – such as schools,
universities, research institutes, etc. On the other hand, politics is often about
debate and disagreement on very delicate issues that determine us as moral beings,
whether we are scientists, artists, or regular citizens. Thus political opinions
often have an inter-subjective aspect that requires justification, debating, and
public criticism. We discuss and try to change each others’ minds about what we
should value and what should be done on issues as diverse as abortion or waging
war, because these issues are important to us as human beings. Thus nobody’s
judgement is taken for granted offhand, because each of us is prone to error due
to the complexity of social reality as well as due to the fact that each of us is a
protagonist in sensitive moral matters that involve value judgements.
Deliberation as a Possible Way-Out

Solving practical social problems in a democratic context means one more
thing: decision reached by democratic process is only a tentative „truth”, and
is also subject to continuous revisions by present and future generations. This
even applies to constitutional decisions. Although we intend to create the articles
of a constitution so to last as long as possible, we cannot see their content as
absolute truth. This should be obvious when we try to interpret the content of
a constitution in order to decide what should be done regarding difficult issues,
issues burdened by our deep value disagreements. Given the fact that democratic
claims stops short of certainty, truth in democracy is not intended to silence
dissenters on the grounds that they have mistaken opinions. The problem of the
minority opinion can be accommodated by the fact that the democratic citizen,
as a member of the majority, should be willing to listen to the other side, who
hold opposing views and make concessions when „forced” to do so by better
arguments coming from the minority. This practice can be developed along the
lines of democratic discourse, the outline of which can be found in Habermas’
famous idea of the „unforced force of a better argument”.11 Although this idea
deserves wider attention as a separate issue, at this point we just want to say
that the idea of deliberative democracy defended here excludes the possibility
of silencing the minority by forcing it into accepting the majority view. In his
work, Habermas develops a promising line on how the claims of normative
rightness can be justified without relying on either the rigid notion of truth
or on the idea of a winning majority in a democratic battle where rival parties
wage a strategic political war. Assuming that the validity of a norm is justified
only intersubjectively by exchanging arguments between individuals, Habermas
11 See his Moral Consciousness and Communicative Action, Cambridge, MIT Press, 1990 p. 160
as well as “Rightness versus Truth: On the Sense of Normative Validity in Moral Judgements
and Norms” in Truth and Justification, Cambridge, MIT Press, 2003.
establishes the standards of „communicative action”,12 which excludes threats,

promises, or any form of strategic action. What counts as a method of settling
practical issues between truthful individuals belongs to mutual understanding
and non-violent persuasion by better argument. The major difference between
instrumental and communicative action is that the former is oriented towards
success (whether in the natural or social world), whereas the latter is oriented
toward reaching understanding in the social world. However, the idea of
democratic discourse understood along the lines of communicative action does
not have to give up on the idea of majority voting. Reaching a decision in politics
can sometimes take a very long time, and if rational consensus stands beyond a
community’s reach, a majority voting procedure may be undertaken in order to
avoid a political stalemate – provided that all available deliberative methods have
been exhausted.13
What structure does a group need to have in order to reach

truth along the „spectrum of objectivity”?
In the case of natural science we shall look to it that the group is composed
of individuals whose level of reliability with respect to truth-tracking is relatively
high (as was mentioned before, most of them should be properly trained, etc.).
Obviously, this group will be less inclusive then the constituency of a polity. The
group of experts in a field should have a coherent system of beliefs as far as
possible and should update these beliefs according to the available evidence. On
average, we would expect that the probability that this kind of group will reach
the verdict that p, when p is true, is very high, and much higher than random.
Of course, this does not have to happen necessarily. The group of experts
might collectively hold an inconsistent set of beliefs: there is no algorithm that
guarantees the consistency of individually-held beliefs at the group level. By
using the majority method of aggregation, even a group of experts can run into
a set of incoherent statements about a certain topic in spite of the fact that each
expert’s individual set of judgements about the same topic was consistent.14
On the other hand, when it comes down to political issues such as capital
punishment or euthanasia, we cannot just rely on experts but must include the
general public for the following liberal reason: since it is a privilege held by a
person to identify and pursue a conception of the good (most liberal thinkers
hold this view. E.g., Mill, Kant, Rawls), democracy has to recognize the ability
of each individual to live in accordance with her own judgement. Assuming that
deliberation can reduce the risk of cycling majorities by making preferences on
a collective level more single-peaked,15 assuming also that we can somehow
12 See The Theory of Communicative Action, Beacon Press, Boston, 1984.
13 Moral Consciousness and Communicative Action, p. 160.
14 See Pettit, P. and List, C., “Aggregating Sets of Judgments: Two Impossibility Results
Compared”, Synthese 140 (2004) 207–235.
15 See Fishkin, Christ, Luskin, McLean, “Deliberation, Single-peakedness, and the Possibility of
Meaningful Democracy: Evidence from Deliberative Polls”, LSE Online: October 2012, http://
eprints.lse.ac.uk/46863/
solve the problem of technicalities associated with deliberative undertaking on

the large scale, we can expect a tendency towards correct collective decision by a
democratic constituency. Democracy is understood here as a form of government
the institutions of which are characterized by both deliberative procedure and
majority rule; deliberative procedure is understood as a collective and mutual
reason-giving activity between citizens who respect each other as free and equal,
with the purpose of reaching a judgement based on reasons to which none could
reasonably reject; where, when and how these two forms of collective decision-
making are to be applied, depends on the context (context depends on the
nature of the issue and on the character and level of the group. For example,
policy advisers, jury in a higher court, nation-state constituency – these could all
require different protocols when it comes to making collective decisions).
3d Aspect: Creating Autonomous Citizens – the

Relationship between the „Market-Place of Ideas” and
Democratic Political Culture
It is probably true that the majority of liberal thinkers view democracy as
a social arrangement that favours autonomous citizens understood as critical
individuals capable of sustaining and reinventing the democratic social context.
This follows from the liberal and democratic ideal of public reason. The idea of
public reason entails that the permissible use of state coercion requires public
justification, which in turn means that coercion can be justified only if it is
acceptable to free and equal citizens on the basis of reasons that they recognize as
valid. The capacity to tell valid from invalid reasons pertains to a critical citizen
prone to reflection about what is good for her as well as about the common good.16
From the institutional point of view, the problem of creating an autonomous
citizen has much to do with how we understand and build the „market-place of
ideas (Mill) in a society. More precisely, this task concerns the problem of what
kind of „flow” of information matters in democracy. For example, is it that „more
things being said” is necessarily better, or is it the quality of the information that
matters? If quality matters, then what is important is not that everyone speaks,
but that everything worth saying is be said. In this sense, ideally speaking, the
market-place of ideas should approximate scientific inquiry and the structure of
institutions associated with them. This structure, as Bernard Williams puts it,
„involves increasingly high entry fees in terms of training, and also, necessarily, a
powerful filter against the crank”.17 But this is a very regulated market. It cannot
be expected from public space.
Nevertheless, the institutional model of public fora in democratic society
should move to a reasonable degree toward an environment that does not
proliferate misinformation, trivialities, scandals, superstitions, or fancies, as well
16 As is well known, the idea of public reason can be found in John Rawls’ work on political
liberalism and public reason, as well as in the work of many other liberal authors working in
this area (Joshua Cohen, for example).
17 Bernard Williams, Truth and Truthfulness, Princeton University Press, Princeton, p. 217.
as other forms of behaviour that contribute to the failure of the proper working
of the market-place of ideas – such as polarization, political silencing, etc. If one
enters a proper democratic market-place of ideas, one should suppose that the
public she addresses is capable of critical reflection. In other words, he or she is
associated with people who are, to use Bernard Williams’ expression, „ready not
to be fooled”, who do not believe everything they hear, and yet who may still retain
great suspicion as to whether there is anything like objective, let alone absolute,
truth in practical matters.18 In this sense one is creating herself as a democratic
individual, capable of sustaining and recreating the democratic environment.
Thus public space should be arranged in such a way that when entering it
one is at pains to convince the public into believing one’s ideas. One’s epistemic
attitude is such that one’s words and ideas are somewhat constrained so that they
can be justified to all. When believing something, one assumes responsibility
for responding to criticism, meeting challenges, and considering objections. In
this sense, when entering democratic public space one is committed to truth
implicitly. The justification of one’s beliefs to the public counts as a part of the
process of truth-acquisition. Obviously this environment requires conditions for
deliberation process, such as mutual exchange of arguments, reciprocity, respects
for interlocutors, etc.19 All this is the matter of political culture, the essential
part of which is institutional setup and content. This brings us to the necessary
institutional conditions for the proper working of the market-place of ideas,
which is probably the most delicate issue for a democratic order. Let us now
consider this problem.
„Machiavellian” Features of a Government20
Transparency is important in any democracy, and especially to deliberative

democracy, where the virtue of truthfulness must come to the fore. However,
as to the virtue of truthfulness, there is a principal discrepancy between the
responsibility of a government on the one hand, and of private individuals on
the other–there is no institutional algorithm that guarantees avoidance of this
conflict. It is true that governments are responsible for the security of their
citizens, and that this obligation cannot always be discharged without secrecy.
Thus government may sometimes withhold information. But from this it does
not follow that it may fool citizens around. The problem is how to control it.
18 Ibid, p. 1.
19 Again, Jurgen Habermas’ project of “communicative action” and “discourse ethics” is of great
help here. His outline of discourse ethics provides for a structure of argumentation that will
yield rational and legitimate answers to moral questions without committing itself to substantive
answers in this domain. When this form of argumentation is applied to the public domain
of deliberative democratic politics, one can say, following Habermas, that laws are legitimate
only if they are arrived at through discourses that are rational, which is to say characterized
by inclusion, equality, and sincerity. See, for more on this, his Between Facts and Norms, MIT
Press, 1996, pp. 107–110, where he discusses the process of validating legal norms.
20 The idea of calling this problem a “Machiavellian feature” of governmental workings was
borrowed from Bernard Williams’ paper “Truth, Politics and Self-Deception”. See his In the
Beginning was the Deed, Princeton University Press, Princeton, 2005, p. 157.
One way to solve this problem is institutional, and is well known from
the work of Montesquieu – this is the idea of a division of political power.
Thus intelligence reports and security secrets that a democratic government
acquires should be shared, perhaps by some members of a legislative part of the
government, or even by opposition party members, or should be available through
parliamentary hearings, etc. Perhaps such institutional practices may reduce the
chance of misinformation, obscuring the truth, and so on. However, this is very
difficult problem for any democracy. What makes it somewhat easier to accept
these pessimistic conclusions is that, institutionally speaking, authoritarian
regimes are less likely to successfully cope with this problem. The problem
becomes even deeper when we think about the issue globally. For example, who
is to scrutinize the work of a government’s intelligence on an international level?
Perhaps this can be done through a set of internationally-recognized institutions,
or non-governmental organizations, or the free global press. The problem with
this is that superpowers, even with a democratic institutional setup, dodge such
issues and very often are not willing to comply to the standards of public reason
in an international context.
Ivan Mladenović Original Scientific Paper
University of Belgrade UDK 321.7:141.7
DELIBERATIVE DEMOCRACY AND THE

EPISTEMOLOGY OF DISAGREEMENT
Abstract:1 Our goal in this article is to explore the relevance of the recent
debate on epistemology of disagreement for democratic theory and especially
for deliberative democracy. When talking about democratic decision making,
deliberative democrats typically stress the importance of consensus or at least
some kind of higher-level agreement. According to one prominent view in the
epistemology of disagreement debate, that might be called the view of conformists,
in the case of peer disagreement we should revise our own beliefs or degrees of
belief so that we end up in agreement with our epistemic peers. It is not surprising
that deliberative democrats welcomed this result. In this paper I shall argue that
deliberative democrats should not embrace too readily the view of conformists.
Keywords: deliberative democracy, epistemic peers, disagreement, the equal
weight view
One of the central assumptions of deliberative democratic theory is that

preferences and beliefs can be changed in the course of a public deliberation.
The problem is that not any change in preferences and beliefs can be considered
justified. The current debate within deliberative democratic theory therefore
largely focuses on the issue whether a change of beliefs can occur in the course
of a public deliberation which would result in bringing truth-tracking collective
decisions (Estlund 2008). Even though early deliberative democratic theory has
been primarily focused on the change of preferences, the center of attention of
current debates are epistemic aspects of public deliberation. In this article, we
will be primarily interested in the epistemic aspect of public deliberation that
concerns the change of beliefs or the degrees of belief. An equally important
proposition of the theory of deliberative democracy is that a change of preferences
and beliefs should lead to a greater degree of agreement among participants
in a public debate and ideally to consensus. However, in political arena with
participants which differ along numerous dimensions, disagreement is more
1 Earlier version of this paper was presented at the Democracy and Truth conference organized
by the Department of Philosophy, Faculty of Philosophy, University of Belgrade, October 5–6,
2013. I am grateful to participants of the conference for very helpful discussions. This paper
was realized as a part of the projects No. 43007 and No. 41004 financed by the Ministry of
Education, Science and Technological Development of the Republic of Serbia.
likely than reaching of a consensus. Our paper will thus attempt to consider
whether it is rational for a participant to change one’s beliefs or degrees of belief
in circumstances of disagreement. Given that it is the central issue of the current
debate on epistemology of disagreement, we will pay special attention to this
debate. Our aim is to explore whether debate on epistemology of disagreement
can help solve difficulties in reaching consensus in the political arena, a problem
faced by contemporary theory of deliberative democracy.
Disagreement is a social fact. Democracy might not be uniquely the best way
to solve the problems of disagreement, but it certainly is normatively, amongst
the most appealing mechanisms. Let’s say that I prefer policy A over policy B,
and that two other persons Mark and Jane have exactly the opposite preference
orderings. We cast our votes and policy B wins. Of course, I still think that this is
not the right decision, that is to say I disagree with them over the outcome of the
decision making process. Still, I prefer to solve our disagreement over policies
A and B by casting votes over using violence or threats of violence. We may
take majority voting procedures to be legitimate even when we disagree with
their outcomes. Some people argued that there is something paradoxical about
this conclusion (Wollheim 1962). I cannot, on pain of incoherence, endorse the
procedure that produces the result that I disagree with.
Deliberative democrats think that if the democratic decision process includes
not only voting, but also public deliberation, then this problem might be solved.
They hope that talking prior to voting may influence some people to change
their views for the right reasons. After hearing reasons for or against different
proposals from other people we may reduce our disagreement over the correct
outcome of the decision making process. If Mark and Jane have good reasons to
support policy B, then I may be convinced by force of the better argument that I
should also vote for policy B. In an ideal case of unanimous agreement, we don’t
need a voting procedure at all, but many political philosophers remain skeptical
about the possibility of reaching consensus in the political arena (Bohman and
Richardson 2009). The most to which we can realistically aspire would be some
kind of meta-agreement (Dryzek and List 2003; Knight and Johnson 2011). Even
if we disagree over our first-order preferences, we may still reach an agreement
over the dimension of our disagreement. That is to say we may agree on what
the common dimension of our disagreement is, and structure our first-order
preferences accordingly. So, deliberative democrats hold that by means of public
deliberation we may arrive at correct or at least at rational social choices. The
democratic decision process aims toward consensus or at least some kind of
meta-level agreement.
It seems that democratic decision making procedures which include
talking prior to voting, have an epistemic advantage over the procedures that
simply aggregate votes or opinions. In order to explain this advantage, Goodin
highlighted the difference between mechanical information-pooling and
discursive information-pooling. Actually, he defines deliberative democracy
as „information pooling by means of talk” (Goodin 2008, 94). Consider the
following example that Goodin uses in order to motivate the difference between
Ivan Mladenović: Deliberative democracy and the epistemology of disagreement 63
mechanical information-pooling and discursive information-pooling. You are

a jury member in a case where a company is being sued for alleged extensive
environmental damage. After hearing all the parties, the foreperson of the jury
asks you to write down two things. Firstly whether or not the defendant is guilty,
and secondly, if guilty, what would be the right amount of money as compensation
of the damage done. You know that the plaintiff asks for compensation of $2
million. Now you think that the defendant is guilty, but you are not sure about the
amount of damages, because $2 million is a lot of money. You decide therefore to
be cautious. You think that $2 million plus or minus half a million dollars is the
right sum, and you are 95% confident that the right sum lies somewhere within
that interval. It is time to write down your opinion. Firstly, you write down that
you find the defendant guilty, but then you cautiously decide to write down the
sum of $1.75 million. After collecting judgments from all the jury members,
a foreperson declares that they have reached a unanimous verdict, because
all members of the jury have written down the same thing. Imagine a slightly
different scenario in which, prior to declaring that the members of the jury have
reached a unanimous agreement, they had a chance to talk about the reasons
that support their judgment. During that process they may learn that they all
were thinking that $2 million is the right sum, but each and every member of
the jury was also thinking that they should proceed cautiously. Upon learning
about the reasoning process of the other members of the jury it may turn out
that „their prudent caution when forming a private judgment is overcome by
confirmation from many other independent and equally reliable jurors” (Goodin
2008, 100–101).
If we take the first scenario as an instance of mechanical information-pooling,
we may easily discover that in the second scenario some extra information is
added that is crucial for reaching the right decision. Goodin claims that „the
discursive process differs in allowing you to probe not only what the other
person’s probability assessment is but also why she thinks that to be so” (Goodin
2008, 95–96). And this is, according to Goodin, the main advantage of discursive
opinion-pooling over mechanical opinion-pooling. Via the discursive opinion-
pooling process we learn not just the probability that other people assign to the
claim that P is true, we also learn whether or not our reporters are reliable. If we
follow the mechanical opinion-pooling process we should revise our probability
assessment that P is true after hearing that some other people believe with much
higher confidence that P is true. As the number of reports that P is true increases,
my probability estimate should increase accordingly, but it is only by means of
the discursive opinion-pooling that I could learn whether my informants are
reliable enough. It may be that Mark and Jane’s higher confidence that P is true
is due to the overconfidence effect, and after learning about their bias to assign
higher probability to events not so probable, the conditional probability that the
proposition ‘P is true, given that Mark and Jane say it is true’, ought to fall. If we
put things in terms of all or nothing belief, rather than probabilistic belief, then
in these kinds of cases we should be completely unmoved by their belief that ‘not
P’, when we believe that P is true.
It seems to me that a deliberative democrat faces the following dilemma.

Call it the dilemma of a deliberative democrat. On the one hand, deliberative
democrats claim that by means of public deliberation we may arrive at consensus
or at least reach some kind of meta-agreement. In any case, public deliberation
should reduce the scope of our disagreement. But on the other hand, in a process
of public deliberation we may also learn about other people’s biases, their weak
arguments etc, which should make us more confident in holding our dissenting
beliefs. Therefore, instead of leading to an agreement, a public deliberation may
end up in some kind of belief polarization. A deliberative democrat cannot claim
that both are at the same time desirable goals of public deliberation. It may be
that deliberative democrats who rely on meta-agreement instead of consensus
have an answer to the aforementioned dilemma. According to those theorists, if
we reach an agreement on a meta-level, it doesn’t matter whether disagreements
over our first-order preferences remain or not. They also collected some evidence
from deliberative polls to support this claim (List et al. 2006). I agree with their
conclusions, but I also think that this answer is ‘preferencist’ in its nature. My
main concern, however, is with the question, should our beliefs or degrees of
belief change when facing disagreement. In what follows I shall explore whether
the recent debate on the epistemology of disagreement might be of some help for
resolving the dilemma of a deliberative democrat.
When we hear the opinion of an expert it might be reasonable for us to defer
completely to their judgment, but what should we do in cases where only partial
deference is warranted? That is to say, what should we do when our advisors are
epistemic peers? When we say that someone is our epistemic peer it usually means
that every piece of evidence that is available to us is also available to that other
person and that we are roughly equal in our intellectual or reasoning abilities (so,
if I am biased in a certain way, then the other people are also likely to be biased
in the same way). I think that deliberative democrats should find this appealing
because they try to make a lot out of the idea of equality. What therefore should
we do when we disagree with someone who is our epistemic peer?
In the epistemology of disagreement literature, there are two main camps
which offer opposing answers to this question. They may be called the view
of conformists and the view of nonconformists (I borrow these labels from
Lackey 2010). I will discuss only the main representatives of these views.
According to the view of nonconformists in cases of peer disagreement, it
might be reasonable for you to stick to your guns if you have good reasons to
do so. Kelly, who is the main proponent of this view, maintains that „once I
have thoroughly scrutinized the available evidence and arguments that bear on
some question, the mere fact that an epistemic peer strongly disagrees with me
about how that question should be answered does not itself tend to undermine
the rationality of my continuing to believe as I do” (Kelly 2005, 170). However,
if we assume there is perfect symmetry between epistemic peers, then there is
no objective criterion that will make it more probable that I am more likely than
my epistemic peer to be correct. As Kelly put it, there is no symmetry breaker.
According to nonconformists the crucial question then is this: „why shouldn’t
I take this difference between us as a relevant difference, one which effectively

breaks the otherwise perfect symmetry?” (Kelly 2005, 179). It is possible, then,
from the point of view of nonconformists, that on some particular occasion one
of the epistemic peers has worked harder in judging the evidence and draws the
right conclusion, and this might be the symmetry breaker. This representative
view that belongs to the family of nonconformist views is usually called the right
reasons view.
Conformists, on the other hand hold, that if we do not have independent
reason to downgrade the opinion of an epistemic peer, then we should revise
our belief or degrees of belief toward that of an epistemic peer. They usually
subscribe to the Uniqueness Thesis. Feldman says that „our options with respect
to any proposition are believing, disbelieving, and suspending judgment. The
Uniqueness Thesis says that, given a body of evidence, one of these attitudes is
the rationally justified one” (Feldman 2007, 205). From the family of conformist
views we will focus on the equal weight view which says that in cases of peer
disagreement we should split the difference, that is, meet our epistemic peers
halfway. Consider the following example (this example is a retelling of the
„restaurant” case from Christensen 2007). Mark and Jane have just finished their
dinner in a restaurant. They have eaten together in restaurants many a time, and
they always split the cost evenly. After seeing the bill, they are both doing the
math in their head and Mark says that their share is $43 each, while Jane says
that their share is $45 each. Of course, they can check their arithmetic again.
But, what if their disagreement remains? Christensen says that in cases like this
„it seems quite clear that I should lower my confidence that my share is $43, and
raise my confidence that it’s $45. In fact, I think...that I should now accord these
two hypotheses roughly equal credence” (Christensen 2007, 193).
When we set the stage for discussing the epistemology of disagreement, we
should note that both views face serious objections. Lackey (2010), who defends
the middle position called the justificationist view, presents these objections in
forms of many against one and one against many. Let’s say that Mark just went to
the Saloma restaurant. He and Jane have a long history of eating dinner in that
restaurant. He meets Jane and says that he is going to Jalan Ampang Street to
have dinner in the Saloma restaurant. Jane then says that the Saloma restaurant
is not in Jalan Ampang Street, but in Jalam Perak Street. Since they’ve been there
many times together Mark might suspect that something is wrong with Jane’s
claim, because he is quite sure that the restaurant is in Jalan Ampang Street. What
if they then meet five other people, who say that the restaurant is in Jalam Perak
Street. Would it be still rational for Mark to have the same degree of confidence
if say a hundred people hold that the restaurant is in Jalam Perak Street. It now
seems counter-intuitive to think that it is reasonable for you to stick to your guns
when many people disagree with you. This is the many against one objection
that targets the view of nonconformists. The view of conformists, however, faces
an objection that might be called the one against many objection. Let’s say that
in the previous example with the restaurant bill there are many other people,
say a hundred of them, who say that the right sum is $43. According to the
view of conformists, despite an agreement with many other people, we should

withhold our belief when we face just one instance of disagreement. This seems
counter-intuitive, too. I will not go any further in answering these objections. In
what follows I want to focus on the equal weight view. My main concern is with
the question of whether or not the proponents of deliberation may rely on the
equal weight view in order to resolve the dilemma of a deliberative democrat.
The reason for this is that some democratic theorists argued recently that the
equal weight view is of the utmost importance for understanding the epistemic
dimension of public deliberation (Peter 2013).
It seems reasonable for a deliberative democrat to adopt the following
strategy. Assume that in a deliberative setting, disagreements are to be expected,
and that people should be treated as epistemic peers. By applying the equal weight
view we may expect that disagreements will dissolve, so by means of the equal
weight view we solve the dilemma of a deliberative democrat. This suggests that
reasonable disagreement is not possible and that in cases of peer disagreement
we should move toward consensus. Fabienne Peter clearly endorses this line of
reasoning when she says that „what has come to be called the Equal Weight View...
claim that reasonable disagreements are not possible. On this view, absent any
further information, the rational response to a disagreement between peers is for
each to meet the other halfway and thus to dissolve the original disagreement.
As long as Uniqueness holds and each has access to the same body of evidence,
two peers who are were likely to perform well cannot both be justified to hold
the original beliefs they do and so a reasonable disagreement between them is
not possible. On this view, rational deliberation between peers forces them to
reach an agreement” (Peter 2013, 1257). But how successful is this strategy? In
order to see that, we now turn to the equal weight view.
Elga provides the following argument in favour of the equal weight view
(Elga 2007, 486–487). Suppose for the sake of argument that it is reasonable for
Mark to be 70% confident that he is correct in judging a horse race between,
say, horse N and horse M, and let’s say that he is confident that horse N won
the race. Now his friend Jane who is as good as Mark in judging horse races is
confident that horse M won. Since Mark is 70% confident that he is right, merely
by noticing disagreement with Jane he got the evidence that he is a better judge
of horse races. Since it is absurd to think that someone is a better judge only
because of noticing disagreements with other people without any independent
reason to think so, the starting assumption cannot be correct. This becomes even
more evident if we take, that instead of one race, there are several races. If Mark
is 70% confident that he is right on each race, after many disagreements with
Jane he will end up extremely confident that he is a better judge of horse races.
Again, it seems absurd to think that someone is a better judge simply because of
disagreements with other people and without any independent reason to think
so. Elga points out that in cases of peer disagreement we should think that other
people are equally likely to be correct (Elga 2007, 487). Assuming otherwise
would lead to absurdity.
Where does this all leave us? Since, „according to the equal weight view, one
should give the same weight to one’s assessment as one gives to the assessment
of those one counts as one’s epistemic peers” (Elga 2007, 484), it seems that it is
possible after all to reach consensus in the circumstances of disagreement. Is it
reasonable then, for a deliberative democrat to rely on the equal weight view?
Here is why the answer is „no”. Elga makes further distinction between pure
and messy real-world cases of disagreement (Elga 2007, 492). The equal weight
view seems correct in the pure cases of peer disagreement, but in the messy real-
world cases it seems absurd because it would recommend suspending judgment
on almost every moral and political question. The problem is that in the pure
cases of peer disagreement, counting someone as an epistemic peer is „based
on reasoning that is independent of the disputed issue” (Elga 2007, 492). That,
however, is not the case in the messy real-world disagreements. Here, reasoning
about the disputed issue is interconnected with many other issues, and there
might be a lot of disagreements about these other issues, too. If we think that
someone got it wrong on many of these issues, then we will not think that they
will get it right on the disputed issue. The upshot is that in the messy real-world
cases we don’t treat each other as epistemic peers. The critical contrast is this:
Think of a smart and well-informed friend who has a basic political
framework diametrically opposed to your own. Imagine that the two of you are
both presented with an unfamiliar and tricky political claim. You haven’t thought
things through yet, and so have no idea what you will eventually decide about the
claim. Still–don’t you think that you are more likely than your friend to correctly
judge the claim, supposing that the two of you end up disagreeing? If so, then
however quick-witted, well-informed, intellectually honest, and thorough you
think your friend is, you do not count her as an epistemic peer with respect to
that claim. And if you do not count her as a peer, the equal weight view does
not require you give her conclusion the same weight as your own. Indeed, if you
think that your friend has been consistently enough mistaken about allied issues,
then the equal weight view requires you to become more confident in your initial
conclusion once you find out that she disagrees.
At the other extreme, think of a smart friend who has a basic political
framework extremely similar to your own. Again, imagine that both of you have
just been presented with an unfamiliar political claim. In this case, you may
well think that in case of disagreement, your friend is just as likely as you to be
correct. If so, and if you and your friend end up coming to opposite verdicts, then
the equal weight view requires you to think it just as likely that she is right as that
you are. But notice that friends like these–friends who agree with you on issues
closely linked to the one in question–will very often agree with you on the one in
question as well. (Elga 2007, 493–494)
The preceding discussion suggests that deliberative democrats should not
embrace too readily the equal weight view. Our central objection is that by
using the equal weight view as a part of the strategy for resolving the dilemma
of a deliberative democrat, one actually assumes what one claims to be proving.
Deliberative democrats think that the common political framework should be
forged through the process of public deliberation. The point here is that in order
to apply the equal weight view they have to presuppose the common political
framework. When this point is appreciated it becomes clear that a deliberative
democrat cannot rely on the equal weight view in order to resolve the dilemma.
Hence the argument based on the equal weight view does not go through.
In this article we have claimed that an important problem faced by
deliberative democratic theory is that public deliberation may lead to reaching
of a consensus, but also to a reasonable disagreement. A deliberative democrat
cannot claim that both are at the same time desirable goals of public deliberation.
We have called this problem the dilemma of a deliberative democrat. The gist of
our article was consideration whether contemporary debate on epistemology of
disagreement can help solve this problem. In that sense, we have particularly
considered the position called the equal weight view which states that in a
situation of a disagreement among epistemic peers, the only rational answer
is meeting halfway, which simultaneously leads to reaching of a consensus.
However, we have concluded that the equal weight view provides only limited
support to the theory of deliberative democracy and that the dilemma of a
deliberative democrat cannot be resolved on its premises.
References
Bohman, James and Henry S. Richardson (2009) „Liberalism, Deliberative
Democracy, and ‘Reasons that All Can Accept’,” The Journal of Political
Philosophy 17: 253–274.
Christensen, David (2007) „Epistemology of Disagreement: The Good News,”
Philosophical Review 116: 187–217.
Dryzek John S. and Christian List (2003) „Social Choice Theory and Deliberative
Democracy: A Reconciliation,” British Journal of Political Science 33: 1–28.
Elga, Adam (2007) „Reflection and Disagreement,” Noûs 41: 478–502.
Estlund, David M. (2008) Democratic Authority: A Philosophical Framework,
Princeton and Oxford: Princeton University Press.
Feldman, Richard (2007) „Reasonable Religious Disagreements,” in Antony L.
(ed.), Philosophers without Gods: Meditations on Atheism and the Secular Life,
Oxford: Oxford University Press, pp. 194–214.
Goodin, Robert E. (2008) Innovating Democracy: Democratic Theory and Practice
After the Deliberative Turn, Oxford: Oxford University Press.
Kelly, Thomas (2005) „The Epistemic Significance of Disagreement,” in Gendler
T. S. and J. Hawthorne (eds.), Oxford Studies in Epistemology, Volume 1,
Oxford: Oxford University Press, pp. 167–196.
Knight, Jack and James Johnson (2011) The Priority of Democracy: Political
Consequences of Pragmatism, Princeton: Princeton University Press.
Lackey, Jennifer (2010) „A Justificationist View of Disagreement’s Epistemic
Significance,” in Haddock A., Millar A., and D. Pritchard (eds.), Social
Epistemology. Oxford: Oxford University Press, pp. 298–325.
List, Christian, Luskin, Robert C., Fishkin, James S., and Iain McLean (2006)
„Deliberation, Single-Peakedness, and the Possibility of Meaningful
Democracy: Evidence from Deliberative Polls,” PSPE working papers, 01–

2006. Department of Government, London School of Economics and Political
Science, London, UK.
Peter, Fabienne (2013) „The Procedural Epistemic Value of Deliberation,”
Synthese 190:1253–1266.
Wollheim, Richard (1962) „A Paradox in the Theory of Democracy,” in Laslett P.
and W. G. Runciman (eds.) Philosophy, Politics and Society, Oxford: Blackwell,
pp. 71–87.
ARTICLES
Jan Narveson Original Scientific Paper

University of Waterloo UDK 321.7:141.7
DEMOCRACY: IS IT BOTH
POSSIBLE AND DESIRABLE?
Abstract: I will be arguing that democracy as classically and normally understood

is no doubt possible, but hardly desirable. And if there is some alternative to the
usual concept that is both recognizably possible and yet desirable, it’ not easy to
say what it is. But a closely related idea might be both – what we might call, the
Democracy of the Market.
Democracy is almost incredibly popular among social theorists and various

popular writers these days. And why not? What’s not to like?
Well, plenty, actually.
Democracy is Rule by The People. The idea is an obvious one. There is Rule
by „the One” such as absolute monarchy; and by „The Few” such as Aristocracy
or the Communist Party; and so by contrast there is rule by The Many, or as we
may as well say, in terms of its presumed ideal, Rule by Everybody.
But of course – to begin with – the idea of Rule by Everybody is riddled
with unclarities, or else is just plain impossible. It would seem to be impossible
if the idea is that we have literally unanimous rule, where each proposed law
is confirmed by a 100% affirmative vote (or disconfirmed by a 100% negative
vote). If we stop short of that, we are driven toward Majority Rule.
Here’s why. More deeply analyzed, Democracy consists in dividing the basic
political power in society equally among all. No one is to have any more than
anyone else. That is its basic idea. If you deviate from that, you necessarily have
rule, not by Everybody, but by Somebody – whoever the designated recipient of
more than equal political power may be.
But what then? How is anything to be done? The most plausible practicable
conceptualization of the equal power that is democracy’s central idea is to convert
this into an equal probability of getting one’s way: the idea is that Your probability
of choosing the rules, or the rulers, is the same as Mine, no matter who You and
I are. This is seen to be maximized at Majority Rule. With more than a majority
required, the Nays have more power than the Yeas. With less than a majority
required, we encounter contradictions – a proposal could both pass and fail at
the same time, leaving us with no result.
So Democracy is Majority Rule. And Majority Rule, at least in the form of rule
by whoever is elected, is certainly at least possible. (How practicable it would be to
have „direct” democracy, even with modern internet techniques, is an interesting
question, but certainly it is not practicable as yet in any sizable community; nor
do most of us think it would be a good idea even if we could do it.)
For that matter, literal majorities are rare in most contemporary democracies.
But if we relax the idea enough to allow election by plurality rather than actual
majority, and make the somewhat generous assumption that the non-voters in a
given election are willing to take whatever they get at the hands of the electorate
who do vote, then government by voting is more than „at least possible” – it is
actually in effect in an enormous number of countries today.
And the results, of course, provide ample evidence that democracy is both
practicably possible and compatible with bad government – very bad indeed.
Well, it is surely worth asking why democracy is supposed to be so good,
anyway?
(1) some enthusiasts seem to think that democracy is an intrinsic good. This
can be analyzed into two variants:
(1a) those who think that Equality is intrinsically good, and Democracy is
one case of that; and
(1b) those who think that, whatever there may be said about Equality in the
abstract, the equal distribution of political power is intrinsically good.
This might be held on the very plausible ground that I don’t want You to
have too much of it over Me, and You don’t want Me to have too much
of it over You – and so, equality is a good bargain, a good compromise.
We’ll say a bit more about that later, pausing only to anticipate further
discussion with the observation that one possible way to divide power equally is
to give everyone none: 0 = 0, after all, and so No power for everybody is Equal
power(lessness) for everybody. Of course, this is no longer what most people
would call „Democracy”; its proper name is Anarchy. About that, there will be
quite a bit more later in this essay.
(2) Perhaps the majority of enthusiasts want to claim that democracy, while
not an intrinsic good, is good because of what it promotes: it promotes, so it
is claimed, the general liberal ideals of society better than alternatives. Those
ideals are, especially: prosperity and peace, but also – insofar as that is different
– the achievement or satisfaction of (other) basic civil rights, such as freedom of
religion, speech, and other such goods. Democracies, they argue, generally are
(2a) wealthier than other societies.
(2b) And as to peace, democracies don’t go to war against each other, and do
find cause to band together against the right sets of bandits and bad guys.
Jan Narveson: Democracy: is it both possible and desirable? 73
(2c) The most important sort of „peace” is the peace of mutual respect for
our liberties including, especially, our Civil liberties: people having their
own religions and lifestyles more generally, without forcible interference
by either their neighbors or by the State.
Since these are empirically checkable claims, we can then go into the facts,
and the results are, I think, at the least mixed.
As to prosperity, there is the developing case of China to worry about. A far
better example is provided by Hong Kong, which has never been a democracy,
but rose from the ashes after the second world war to become one of the
wealthiest societies in the orient. What’s more, it is very doubtful that it would
have done so well if it had been a Democracy. Obviously democracy is neither
a necessary nor sufficient for wealth. The most that can be said is that most
quite wealthy countries are democracies, and only a few wealthy countries are
not democracies. So the connection between democracy and wealth is at least
loose. This is not very surprising, for reasons well known to economists.
As to international peace, the case that democracies don’t make war on each
other is pretty good, though we get into niceties of definition after awhile. On
the other hand, there is the suspicion that the world is as peaceful as it is (ahem!)
because of the long nuclear hegemony of the two „greatest powers,” especially of
the U.S.A.
Unfortunately, the U.S.A. stands out, along with many others, as an example
of how civil and other liberties can suffer at the hands of democracy.
Let’s turn back to theory for a moment. Let’s first examine the pure idea
of majority rule. Some would deny that democracy is majority rule. How you
explain the basic idea of „rule by the people” without that idea is an interesting
question, to which we’ll return briefly below. But let’s start by just considering
majority rule – „mob rule” if you like.
It quickly becomes obvious that a majority could violate any number of
liberal desiderata. A majority could impose religions, suppress freedom of
speech, make war on homosexuals or drug users, impose onerously high taxes
and otherwise mismanage the economy. Come to think, it not only could do all
those things, it has done them, and in spades. The point is that liberalism is one
thing, democracy is another, and there is considerable tension between them.
All democracies today have constitutions, limiting to some degree, somehow,
the potential of majority rule for curtailing individuals liberties. It is often said,
indeed, that „If various non-majority rights are not upheld, then it simply isn’t a
democracy.” Unfortunately, how you establish a conceptual connection between
the two isn’t gone into: the claim is simply that Democracy = elections plus
constitutional constraints –– no questions asked why, if majority rule is so great,
we should have individual rights to worry about at all, nor what democracy
has to do with such rights. The point of this essay, though, is to push that very
question – hard.
There are two, though related, general problems about democracy.

1. The first is the constituency problem: who is the public that democracy
applies to? We have various polities: cities, provinces, countries. Without
thinking about it, we assume that democracy applies to whatever polity
we are talking about. This unthinking acceptance, as we’ll see shortly,
leads to immense problems.
2. The other is the agenda problem: just what may the democracy do?
What may voters vote about? What may elected politicians do with
their conferred power? There is a laudable tradition having it that
government is or should be for the public good. While that is widely
regarded as an extremely vague or perhaps vacuous idea, it is not –
once we accept liberalism as our general outlook. But even within the
constraints of liberalism, to be sure, there has been disagreement about
this. While I think we can narrow that down, what is not obvious, as we
will again see, is that democracy can be expected to work very well for
generally promoting that good. Indeed, the more clear and relevant our
refinement of the appropriate notion of public good, the less obvious is
it that we could expect democratic decision making to do much for it.
1. The Constituency Problem

If democracy governs a particular constituency, why does it? What business
do some of the citizens have running the lives of others on, say, the other side of
the country or the town? Or if it does, why doesn’t it also apply to people on the
other side of the world? Let’s draw a great big arbitrarily formed line enclosing
an arbitrarily large set of people, here and there and having, as we would
normally think, pretty much nothing to do with each other. Now the question is:
if democracy is supposed to be some sort of basic right, why don’t all the people
in that area, whatever it is, get ruled over by majorities within that same area?
For that matter, why isn’t everybody on earth to be ruled over by a majority of
mankind at large?
There is, to be sure, a common sense answer to this: it’s that those people
have, as we might say, „nothing to do with each other” – they have „no business”
interfering in each other’s lives. And that is indeed true! But the question is – if
Democracy is so great, why is that relevant? Given democracy’s pretensions to
the status of a basic right, how can this commonsense point, that the people over
here may have little in common or to do with the people over there constitute a
reason why either or both shouldn’t be subject to the rule of the others? Why is
it so obvious that people in an arbitrarily defined area of the globe should not be
subject to majority rule about anything, if democracy is worth its salt?
The moral here is that first we must know why these and just these people
are „together” in such a way that it makes some sort of sense for them – them
and not the umpty-million other people around the world – to perhaps resolve
some mutual problems by voting. Whatever it is, it is clear that such a theory
would have to be prior to democracy.
Consider this, from Melissa Schwarzenberg:1 „The reason why we ought to
want democratic decision making in the first place is because we properly believe
each member’s views and opinions ought equally to affect the outcome of group
decision making.” That is her version of an answer to my question. But what does
she mean here? She could mean either that we, as it were, always do „properly
believe” – that it is proper to believe – that each person’s views ought equally etc.?
Or does she mean that if in a given case it is (for some reason, not yet stated)
„proper” to believe that, then we should want democratic decision making?
The difference is radical, though I am not sure that Dr. Schwarzenberg has
noticed that. Enthusiasts for democracy, I rather think, just tend to think that of
course just everything is such that everybody ought to have an equal say about it.
But ordinary people believe no such thing, we may be sure, and it is hard to see
any reason why they should. I do not think that the opinions of my fellow man
are in the remotest way wanted or to be consulted, let alone given „equal weight”
with, say, those of my wife when it comes to doing the grocery shopping, or
deciding whether to take that job in Connecticut, and so on. I don’t think that,
and I doubt that much of anyone else does either. You’d have to be a very „far-
out” ideological enthusiast to think any such thing.
Where, then, is the argument needed to explain why just this and only this
sizable group of people ought to be democratically equipped for decision-making
about this or that issue for that whole particular group?
2. The Agenda Problem

So, which things are suitable for democratic decision-making, and, of course,
why? Here’s my sample whipping-boy issue: should we crucify Jim Smith in the
city tomorrow, just to watch the spectacle? Would the approval of a majority
make it right – should the „mob” indeed „rule”? When we recoil at that, the
question is, why we aren’t recoiling at the very idea of democracy? Most people
don’t think we are, of course, but the question is; why?
What „we” think, surely, is that we should not be able to vote on such an
issue, at all. Whatever democracy is about, it surely isn’t that. What we think,
and with extremely good reason, is that there are things nobody is permitted to
do to anybody, whether that person is in some government or not, and whether
a whole lot of other people are in favor of it or not. These are rights. Most of us,
I trust, think that people do have them. To hold this is not to be bizarrely right-
wing, or some such silly category. It is, surely, common sense. Governments,
mafiosa, your next-door neighbor – you name it – for those people to do such
things is wrong. It’s also wrong for any majority to do them. That isn’t because a
majority has decided that it’s not suitable for majority rule. It’s because such an
issue is not suitable for majority rule, and so no majority should think it is.
1 Melissa Schwarzenberg, Counting the Many [Cambridge U. Press, NY, 2014], p.116.
And so a decent „democratic” society will have constitutional restrictions

on things like that. Just which things are „things like that” is, of course, the
crucial question. And which things are properly on the „agenda” for democratic
decision-making?
The democratic tsunami begins to dissipate when we seriously contemplate
these last questions. Some constitutional restraints can be defended in the
interests of democracy, to be sure. For democracy to function, the right to vote
cannot be based on majority rule. Similarly, there must be freedom of the press,
at least for political reporting and communicating. Elections must be fairly
frequent. And one civil liberty (sort of) is surely called for by democracy: people
must be free to run for office, and free of intimidation for political opinions.
Legal rights, such as to avail oneself of legal services, are in the same category.
We might, then, distinguish between all-out, instantaneous democracy, and
democracy as an enduring political system. It is democracy of the latter kind that
the term ‘democracy’ is usually applied to.
But while those specifically political rights are essential to enduring
democracy, they are, after all, far from all there is. Freedom of religion, and
broadly speaking of lifestyle subject only to the requirement of nonviolence, are
not obviously required for democracy to function. As noted above, it is all to
easy to envisage an enthusiastic majority supressing the religious aspirations of
some smaller group. What’s to prevent it? And indeed, democracies for as long
as there have been such have indeed to greater or lesser extent infringed, mildly
or severely, just such rights.
And then, perhaps above all, there is freedom of commerce and the right to
earn income and own property. Now, these are generally regarded as contentious,
and rightly so. For if we acknowledge a strong right of property, then the state
has problems, since essentially everything the state does rests on taxation, or its
equivalent, or some other use of coercion. No taxation (or the equivalent), no
government. And yet, taxation on the face of it looks to be an infringement on
that kind of freedom. Taxation says: you can’t spend (what you thought was)
your money as you like – you must spend it on this and that – more generally, on
what we, the elected government, tell you to.
Now, ordinary people don’t think that in order to find out what I should do
in life or how I should spend my money, we should take a vote. These things are
ordinarily thought to be my business – the business of the person whose life it is,
or whose money it is.
What other people might be relevant? Clearly there are often others: the
people who pay me for the services I render to earn my money, certainly; and
my wife, quite likely; and possibly a few others, or even many others. With all
of those people, however, we proceed by making arrangements. Those persons
own various things, can provide various services to me; I own various things and
can provide various services to them. So we arrange exchanges, in many cases.
No problem! (That is, no fundamental problem, however subtle and intricate the
structure arrived at, or the negotiations leading to the arrangement in question.)
We have, in other words, a mechanism for making decisions that is obviously

completely appropriate and called for, and equally obviously does not involve
most people, let alone all people. Majorities are irrelevant to such things. And
such things, after all, are at least a great deal of anybody’s life.
When, then, is a political procedure such as democracy relevant? What,
so to say, is the natural agenda for politics, hence for democracy? Here again
there is a familiar and perhaps potentially decent answer: „Public” goods, goods
that matter to the „everybody” who form the potential constituency in question.
The name implies that the public is relevant to those matters. And that there are
different publics for different matters is very obvious. We may seem to have the
beginnings of an answer to our questions.
But to see the exact bearing of the notion on our question, we need a more
precisely characterization of publicness. The standard such one is, roughly, that
a public good is one whose production and consumption are such that it is
impossible, or at least very difficult, for one person to be involved without others
being so too. At a minimum, it’s a good that one person cannot enjoy, or produce,
unless others do too: the person who produces it cannot prevent others from
enjoying it without, in effect, paying for it. Or, it is possible to enjoy it without
paying the producer for it. In consequence, there are obvious incentive problems
about their production. If producers can’t control their markets, they won’t be
inclined to produce these goods. This is known as „market failure.” So, now, the
idea goes: the natural thing is to tax consumers so that they have no choice about
paying. And nobody can tax everybody without governmental powers.
But then we have at least two problems. The first, obviously, is that some
persons won’t think that the extracted price is worth it. If they had their choice
between paying and receiving, or not receiving and not paying, they would
choose not to pay. Is the cure worse than the disease? Frequently, the answer
is – definitely!
Now, the democratic mechanism will let this be decided by voting. And
this gives rise to the problem, first, of misperception of the costs and benefits of
various proposed provisions of what are allegedly public goods. Secondly, unless
there is some kind of filter or constraint on the range of issues that can be treated
in this way, we are back to all of the original problems with democracy. This is
the problem of government failure: government action that defeats, rather than
promotes, the good of its people.
Government fails when it provides what people don’t want or provides what
they do want, but at a price they’d be unwilling to pay given their choice; or
when it fails to provide what they do want at prices they would be willing to
pay, if they had their choice. Government characteristically denies them these
choices. It imposes a school tax on all, and provides schooling that many think
unsatisfactory. Those people might not be allowed to use private schools; or if
they are allowed to, those who avail themselves of such schools will be paying for
the public ones anyway, and so their costs for schooling are much increased. And
so on, for almost every other „public good” that governments claim to provide.
A way of putting this difficulty is to note that the constituency of different

public goods varies, enormously. That is the constituency problem, again. Ideally,
the constituency would be all and only those who care about and are willing to
participate in the creation of the goods in question.
How, ideally, would we do this? Let’s start by going back to the notion of
public goods. It has been often enough pointed out that many of the areas in
which government is extensively concerned are not in fact public goods, in the
technical sense, at all. Consider, for example, health, education, and welfare
– three areas that use up by far the lion’s share of most contemporary public
budgets. Yet none of them is a „public good”! All of them could be and to a
considerable extent are provided without government involvement.
1. Health mostly is not a public good. I can be sick or die, independently
of whether you are or do. There is a small subset of genuinely public-
health problems: infectious diseases, especially. Most health problems
are not due to infectious diseases. And even ones that are can be
resolved, usually, by quarantine. Any liberal would have to concede the
justifiability of quarantine for a small range of public health issues. But
that leaves almost all health issues non-public. Yet in most countries
today, it is simply assumed that your health is automatically to be
regarded as part of the public’s proper area of concern. In Canada, we are
not even allowed to buy private medical procedures (unless we leave the
country and get them elsewhere.) People endure months and months of
waiting for care that would be all but instantaneous south of our border,
for those who can pay. (It will be asked – it always is, incessantly, asked:
what about those who can’t? We’ll discuss that in a bit.)
2. And welfare is not obviously a public good, either. One person can be
well off, another destitute. Indeed, it is not only „not obviously” a public
good – it is, rather, obviously not a public good.
What is true is that if person A is destitute, the price he can get for
whatever services he may be able to perform will be very low. In
consequence, one would in the normal course expect destitution to be
very rare, in any society in which people are allowed freely to buy and
sell, or beg for, services. Contemporary democracies, however, rarely
allow this. Long before destitution, people are provided with various
welfare services – at taxpayer expense. And they provide those well
above the level we can plausibly regard as that of „destitution.”
Many people will think it immoral not to do this. Yet few people think
it’s moral for A to feed B by robbing C – which is exactly what the tax-
supported welfare system apparently does. A long-standing problem in
democratic theory is: why isn’t it robbery? Why isn’t government a den
of thieves? It needs a decent answer. If welfare were genuinely a public
good, that would provide the „decent answer” called for. If it were
impossible for me to be well off unless B was also well off, then anyone
who wants to be well off (and who doesn’t?) would have to accept that
we must see to B’s welfare when we pursue our own. But of course, it’s
not impossible. And then what?
3. As to education, it is obvious from the start that A can be educated and
B not – there is no obvious reason to think that if anybody is educated,
then everybody must be. or vice versa, or that if A educates B, it will be
impossible to prevent C from horning in for free.
It has become a mantra of contemporary societies that people have,
somehow, rights to the (free!) provision of all these things. Well, why? Of course,
it can also be pointed out that it is not so long ago that no such thing was
thought by most people. Governments did not think it their duty to provide any
of the above. What they may have thought it their duty to provide, and what
their constituents very likely thought it their duty to provide, was security and,
hopefully, liberty – two things that governments have been notoriously erratic
about, to say the least. And those are indeed public goods. I cannot be secure
unless you somehow refrain from assaulting me, and vice versa. I can do x only
if you don’t prevent me, and vice versa. Liberty is liberty from the incursions
of others; security is the absence of danger from others. Only action (that is,
inaction, especially) by others can provide these thing.
Majorities can indeed act so as to help defend, and even improve, the security
and liberty or each of us. Unfortunately, they can also do just the opposite, as we
have seen. Since democracy gives every citizen power over every other citizen,
in the form of the vote, the problem is how to restrict it, to tie its hands, so that
it will tend to promote rather than defeat the security and liberty of each of us.
Democratic Voting
Why do people vote, when they do? That is, what are they trying to achieve
by voting? There are two general answers to distinguish here.
(1) Sometimes, they are trying to promote their own interests, but by
political means.
(2) In other cases, they have ideas about what society should be like –
what they want other people to be doing. Those are akin to political
philosophies, more or less (mostly less, of course, in the sense of
articulated and developed doctrines, but nonetheless political attitudes
rather than expressions of, strictly speaking, self-interest.)
For our purposes, we need not try to make more refined distinctions, as for
instance between „strictly speaking” self interest, interests in family, in ethnic
associates, in religious affiliates, business partners, and so on up to national
and perhaps global sympathies. What does matter is that both of these general
categories, and all those in between, raise the same difficult question: namely,
why? In the case of self-interest, why should Harry be subject to coercion in
order to benefit Hilary? – and in the case of ideological-type motivations, why
should Harry be subject to coercion in order to advance a political philosophy
held and pressed by Hilary, but with which he, Harry, does not agree?
This problem is perfectly general. No matter where in the spectrum our

motivations lie, the trouble is that others will be acting out of different ones.
And since what makes our actions rational is that they are reasonably expected
to promote whatever sort of interests they are motivated by, it is on the face of
it extremely unobvious why we should be subordinated to somebody else’s. Make
this same point regarding democratically elected people whom one didn’t and
wouldn’t vote for, and you have the problem about democracy in a nutshell.
It may be said that these people have authority over you. But the question
is, why do they? Enthusiasts for democracy, after all, say that they have it because
you gave it to them! But the trouble is, you didn’t. And even in the case where
you did vote for the chap, it is quite likely because you felt he wouldn’t be as
horrible as the other chap running for that office – rather than because you felt
he would do a good job of it.
Liberals think – rightly in my view – that presidents and congresspersons
and such who are voted into office have the democratic merit that we, the
people, put them there, and thus endowed them with the authority attached to
those particular offices. But the trouble with that story is that it is, almost always,
false, either in general or in detail, when we try to recast the story at the level
of the individual – who, after all, is supposed to be the fundamental element
in the story. „We” the people did not „put them there” – about all we did, and
do, is, to go to the polls (perhaps – for many of us do not), select some name
scarcely known to us, almost all of whose expected actions are unknown and
unpredictable, and then stay out of their way if we can. The standard democratic
story about elected government doesn’t bear scrutiny.
It has been widely noted that the public is relatively susceptible to „campaign
promises” which are not really expected to be kept, especially because people
have a vague appreciation that they really can’t be. We can’t, for example, have
higher government spending, lower taxes, and a reduced deficit simultaneously.
The sky is not good at supplying us with pie. Two and two still do make four,
despite the machinations of politicians.
The question is: Why are dishonest promises so often successful? And,
should politicians be able to make such promises? Is there any way to make it at
least difficult for candidates for public office to get elected, and re-elected, on the
basis of knowably and known false claims?
Public Goods, again

We might be able to help matters out if we could design a constitution with
filtering mechanisms that forced governments to confine their legislation and
administration, in the first place, to genuinely public goods only – goods that
really cannot be taken care of by market methods – and second, to ones where
the stakes are nontrivial. Neither, to put it mildly, is very easy.
3. Superiority of Markets
As to the first, we should surely have a constitutional mechanism for assuring
that anything that can be done by the market should be done that way, rather
than by a political method. Markets have the signal virtue that since nobody has
to buy or sell, they target goods at those interested, and not those who are not.
The fundamental problem guiding this inquiry is how to justify imposition by
majorities on minorities, that being the defining hallmark of democracy. The
market avoids this, because all parties are required to respect the property rights
of all persons. (That will only be an adequate characterization if we understand,
as Locke did, that one’s „property” includes one’s own actions, body, and
psychology. What would usually be called ‘property rights’ specifically would be
extensions of the basic rights that Locke specifies, to life, health, and liberty –
all such rights being negatively construed, as Locke does in pronouncing the
fundamental „Law of Nature” to consist in a prohibition of harm of all those
things, plus property in the narrow sense.)
This outcome would be assured by strong property rights. If we cannot,
as John Locke’s idea would have it, impose taxes against the consent of each
individual taxed, then confinement to market methods is evidently assured.
This might be thought elementary – and we can also be confident that such
a provision has no real-world chance of being incorporated into any constitution.
The interesting question is, Why not? And the answer, I think, has to be, serious,
near-total, misunderstandings of the free market, as an institution and as an idea.
Alas, that is the subject of books, and calls for a much longer treatment than this.
4. Public Goods: The Variables

When public goods are loosely characterized as goods such that either
nobody has them or everybody does, or as goods which it is „impossible” to
prevent free-riders and so on, we may just close the subject by pointing out that
there are very few if any goods of that kind. There is one good, however, that is
literally impossible to have without the cooperation of at least two persons. This,
as pointed out above, is the good of Peace. Jones is at peace only if Smith refrains
from attacking him. The world enjoys peace only if nobody makes war. With any
group X, X is at peace internally only if no one in X makes war against anyone
else in X. And X is at peace externally only if the Xians, as a people or a State, are
not at war with any other group Y.
Beyond that, things get more difficult. What must happen is that the
producers of a putatively public good must decide how much cost they will
accept in order to confine their products to those who will pay. The producers of
goods with externalities, such as pollution, will be subject to tort actions by those
affected, unless they are able to make a deal with the involuntary consumers that
will turn them into voluntary ones.
When such matters are left to the democratic State, we may be sure that
things will go wrong, mostly in the direction of oversupply: more of the „good”
will be supplied than people would have been willing to buy at that price if they
had their choice.
Roughly, we may say that public goods are supplied mostly at some cost to
peace: some people are coerced, compelled to endure unwanted expenditures, in
order to enable provision. And if it has to be coerced, we may conjecture that it

can hardly be a literally public good.
Where there’s room for The Public

Now and then there’ll be a public good that reasonably well exhibits an
important game-theoretic profile: the coordination dilemma. While ‘dilemma’ is
a misleading expression, since most are not the sort that call for much agonizing
over, it is still accurate in that we have a decision problem with, in a sense, no
obvious solution. Coordination dilemmas are cases where (1) it is in „everyone’s”
(any set of at least two persons) interest to adopt the same strategy as everyone
else regarding the matter in question, but (2) there are alternative strategies
no one of which is superior to some or all others, and thus if people choose
independently, we will almost certainly come up with a decidedly suboptimal
solution. The classic example is the Rule of the Road: we keep right because
everyone else does, and if we all do, we all benefit relative to the no-rule situation,
which we presume to be the base line. If some go right and some left, we will
have squealing brakes, curses, and crashes. If all drive on the same side, on the
other hand, then we all arrive safely at our destinations, barring other problems.
We can solve the which-side-of-the-road problem easily in practice. But
what if there are a dozen alternatives, equally good in themselves? What if some
are preferred by some and others by others, even though they all get the some
benefits for all? At last! – a real use for Government! (Maybe.) And, since it
really doesn’t matter, why not go with the majority? Voila! An argument for
majority rule, at last! But then we come back down to earth when we appreciate
how very, very few problems of this type we are faced with. (And also, we went
too quickly: the potential for solving multi-dimensional coordination problems
without a Central Director is at least considerable.)
Insofar as politics works on this sort of coordination problem, we perhaps
arrive at our Window where democracy looks like a good method of solving
it. Where it doesn’t really matter, going with the majority entails no losses, and
some gains. So, why not?
The question would be, though, how often this applies. Politics, it seems
to me, is not very often such a problem, and when it’s not, as is much more
frequent, then democracy, like other sorts of government, is guilty of sacrificing
minorities to majorities (and both, of course, to pressure groups and lobbyists).
Governments strongly tend to revert to their standard function of „misdirecting
massive resources from where they are needed to where politicians want them to
be.”2 They will over-regulate, over-punish, over-rule, and just generally make life
less desirable than it might be, and, alas, Democracy does not help very much.
Is a society that operates fully on the market system possible? That’s an
interesting question, to which the standard answer is roundly in the negative.
But why? A suggestion of the reason lies in this interesting characterization
2 I owe this pity summary to Henry Hazlitt, author of the essential book Economics in One Lesson.
of democracy, a saying of a long-ago colleague: „Democracy only settles the

question who would win in a fair fight!” Politics is about who is going to exert
power over us, and not how to control, contain, and limit that power.
It is often said, famously by Churchill and many others, that democracy,
whatever its faults, is at any rate better than any other form of government.
Whether that is even slightly true depends on what you call a „form” or „system”
of government. No democracy today is a pure democracy in the sense that it is
unconstrained exercise of the „public will” – majority rule,”mob” rule. They differ
greatly among themselves on just which patterns of restrictions they impose, but
broadly speaking, there tends to be a fairly substantial component of Liberalism
– of respect for individual liberties, of embracing the general goal of promoting
the individual good of each citizen – in almost all of them. A fascinating question
is whether we can reduce the component of majority („mob”) rule – to zero, and
allow individual liberty to „rule” exclusively. Let’s at least hope that mankind can
do better than we mostly do nowadays.
Nicholas Rescher Original Scientific Paper
University of Pittsburgh UDK 165.4
167.33:165.4
OVERSIMPLIFICATION
1. Why Natural Science Oversimplifies Matters

Simplification can of course arise either knowingly and deliberately, or
unknowingly and inadvertently. Oversimplification–simplifying matters beyond
the warrant of the functional requisites of the particular situation at hand–
is almost always inadvertent. Usually the only context in which it is done
deliberately is in instructional situations where it function as a heuristic way
station to more adequate subsequent treatment.
Inadvertent oversimplification occurs because we are not omniscient.
In the final analysis oversimplification is inevitable for limited intelligences
seeking to come to grips cognitively with an endlessly complex world. As beings
whose actions are guided by thought we constantly have questions that require
to be answered in real time, in circumstances where acquiring and processing
the requisite information simply takes too long. To get from where we are to
where we need to be demands shortcuts across an informational vacuum. And
oversimplification is the only way to manage this.
Oversimplification of the real is inherent in the very nature of cognitive
rationality as it functions in scientific inquiry. For empirical science is a
matter of drawing universal conclusion („theories” they are usually called)
from the perceived facts of observation and experiment. But observation and
experimentation is ongoingly enhanced by technological advance in the devices
used to monitor and manipulate nature. And our theories fit the existing data
tightly. And so the web of theory that is woven about a given manifold of data will
not–and effectively cannot–be adequate to the situation that obtains subsequently,
after our body of information has become enhanced. It is–inevitably–oversimple.
And so as our data are amplified through new observations and experiments,
the previously prevailing theories will almost invariably become destabilized.
Those old theories oversimplified matters: new conditions call for new measures,
new data for more complex theories. It lies in the rational economy of sensible
inquiry that the history of science is an ongoing litany of oversimple old theories
giving way to more sophisticated new ones that correct their oversimplification
of the old.
In natural science we ongoingly enlarge the window through which we
view the phenomena of nature. As we build more powerful telescopes or
microscopes, more elaborate particle accelerators, more sophisticated apparatus
for low temperature experimentation closer to absolute zero, etc. we bring new
phenomena to view. And new phenomena generally do not fit into old patterns.
There is no fact about the history of science that is established more

decidedly than this: that new technology (be it material or conceptual) puts new
data at our disposal and that new data manifest the oversimplification of earlier
theories.
We package our account of nature’s phenomena into the smallest and most
compact explanatory box that we can manage to contrive for them. And as the
scientific progress enhances our means for experimentation and observation it
enlarges the volume of data at our disposal. The old package no longer manages
to encompass the whole. We must go back to the old drawing board to construct
a larger explanatory container. And this will complicate the situation. The former
state of the art will, in respect, seem oversimple.
The reality of it is that in scientific theorizing we proceed along the lines
of least resistance, seeking to economize our cognitive effort by using the most
direct workable means to our ends. Whenever possible, we analogize the present
case to other similar ones, because the introduction of new patterns complicates
our cognitive repertoire. And we use the least cumbersome viable formulations
because they are easier to remember and more convenient to use. Science tries
to transact explanatory business into the simplest, most economical way: In
explanation as elsewhere it is only rational to adopt the simplest solutions. But
the new phenomena destabilizes the old simplicities and call for more complex
responses. But novelty produces complexity. In every branch of science, today’s
handbooks, and manuals are thicker than yesterdays. From the point of view of
the present, yesterday’s science always seems oversimplified.
The fact of it is that scientific progress is a matter of correcting for
oversimplification because over-simple theories invariably prove untenable in a
complex world. The natural dialectic of scientific inquiry ongoingly impels us
into ever deeper levels of sophistication.1
Aristotelian mechanism did not distinguish between weight and mass.
Newtonian physics did not distinguish inertial from gravitational mass. Cartesian
dynamics did not distinguish between force and action. Mendeleev’s chemistry
did not distinguish atoms from their isotopes. And so it goes. Enhanced
observations evolve more elaborate theories that require more sophisticated
distinctions. From the vantage point of later science the earlier efforts look to be
oversimple.
The ancient Greeks had four elements; in the nineteenth century Mendeleev
had some sixty; by the 1900’s this had gone to eighty, and nowadays we have
a vast series of elemental stability states. Aristotle’s cosmos had only spheres;
Ptolemy’s added epicycles; ours has a virtually endless proliferation of complex
orbits that only supercomputers can approximate. Greek science was contained
on a single shelf of books; that of the Newtonian age required a roomful; ours
requires vast storage structures filled not only with books and journals but
with photographs, tapes, floppy disks, and so on. Of the quantities currently
recognized as the fundamental constants of physics, only one was contemplated
1 On the structure of dialectical reasoning see the author’s Dialectics (Albany NY: State
University of New York Press, 1977), and for the analogous role of such reasoning in
philosophy see The Strife of Systems (Pittsburgh: University of Pittsburgh Press, 1985).
Nicholas Rescher : Oversimplifcation 87
in Newton’s physics: the universal gravitational constant. A second was added in

the nineteenth century, Avogadro’s constant. The remaining six are all creatures
of twentieth century physics: the speed of light (the velocity of electromagnetic
radiation in free space), the elementary charge, the rest mass of the electron, the
rest mass of the proton, Planck’s constant, and Boltzmann’s constant.2
The taxonomy of physics provides a further illustration. In the 11th (1911)
edition of the Encyclopedia Britannica, physics is described as a discipline
composed of 9 constituent branches (e.g., „Acoustics” or „Electricity and
Magnetism”) which were themselves partitioned into 20 further specialties (e.g.,
„Thermo-electricity: of „Celestial Mechanics”). The 15th (1974) version of the
Britannica divides physics into 12 branches whose subfields are–seemingly–too
numerous for listing. (However the 14th 1960’s edition carried a special article
entitled „Physics, Articles on” which surveyed more than 130 special topics in the
field.) When the National Science Foundation launched its inventory of physical
specialties with the National Register of Scientific and Technical Personnel in
1954, it divided physics into 12 areas with 90 specialties. By 1970 these figures
had increased to 16 and 210, respectively. And the process continues unabated
to the point where nowadays people are increasingly reluctant to embark on this
classifying project at all.
Oversimplification here occurs because we are not omniscient. The root
cause of oversimplification is ignorance: we oversimplify when there are features
of the matters at issue about which we lack yet-unavailable information. And it is
somewhere between hard and impossible to come to terms with this. Ignorance
is the result of missing information, and one level of this is inevitably tenuous.
What is at work here is one of the fundamental principles of epistemology:
We are bound to be ignorant regarding the details of our ignorance. I know that
there are facts about which I am ignorant, but I cannot possibly know what
they are. For to know what such-and-such is a fact about which I am ignorant,
I would have to know that this is a fact–which by hypothesis is something that
I do not know. And the same situation prevails on a larger scale. We can know
that in various respects the science of the present moment is incomplete–that
there are facts about the working of nature that it does not know. But of course I
cannot tell you what they are.
An inherent impetus towards greater complexity pervades the entire realm
of human creative effort. We find it in art; we find it in technology; and we
certainly find it in the cognitive domain as well.3 And so we have no alternative
to deeming science-as-we-have-it to afford an oversimplified model of reality.
And in consequence science cannot but enmeshed in the same shortcomings
that beset oversimplification in general.
2 See B. W. Petley, The Fundamental Physical Constants and the Frontiers of Measurement
(Bristol and Boston: Hilger, 1985).
3 An interesting illustration of the extent to which lessons in the school of bitter experience
have accustomed us to expect complexity is provided by the contrast between the pairs:
rudimentary/nuanced; unsophisticated/sophisticated; plain/ elaborate; simple/intricate. Note
that in each case the second, complexity-reflective alternative has a distinctly more positive
(or less negative) connotation than its opposite counterpart.
Why do we ever oversimplify? Why don’t we simply take those ignored

complications into account? The answer is that in the circumstances we simply
do not know how to do so. The situation is akin to that of the Paradox of the
Preface. Recall that here the author writes:
I want to thank X., Y, and Z for their help with the material in the book. I
apologize to the reader for the remaining errors, which are entirely mine.
One is, of course, tempted to object: „You silly author! Why apologize for
those errors? Why not simply correct them?” But alas the reader just does not
know how. That there are errors he realizes; what they are he does not. And
the situation with oversimplification is much the same. All too often we realize
that we oversimplify, what we do not know is where we oversimplify. This is, in
general, something that we can discern only within the wisdom of hindsight.
2. The Price of Oversimplification:

Cognitive Myopia
Oversimplification becomes a serious cognitive impediment by failing to take
note of factors that are germane to the matters at hand, thereby doing damage to
our grasp of the reality of things. Whenever we unwittingly oversimplify matters
we have a blind-spot where some facet of reality is concealed from our view.
Oversimplification thus consists in the omission of detail in a way that is
misleading in creating or inviting a wrong impression in some significant–i.e.,
issue-relevant–regard. In practice the line between beneficial simplification
and harmful over-simplification is not easy to draw. Often as not it can only be
discerned with the wisdom of retrospective hindsight. For whether that loss of
detail has negative consequences and repercussions is generally not clear until
after a good many returns are in.
For the most part, oversimplification involves loss. The student who never
progresses from Lamb’s Tales from Shakespeare to the works of the bard of Avon
himself pays a price not just in detail of information but in the comprehension
of significance. And the student who substitutes the Cliff ’s Notes version for
the work itself suffers a comparable impoverishment. To oversimplify a work of
literature is to miss much of its very point. Whenever we oversimplify matters
by neglecting potentially relevant detail we succumb to the flaw of superficiality.
Our understanding of matters then lack depth and thereby compromises its
cogency. But this is not the worst of it.
Oversimplification always involves errors of omission. It occurs whenever
someone leaves out of account features of an item that bear upon a correct
understanding of its nature. For example, to say that Rome declined because
its elite was enervated by lead poisoning from the pipes of its water supply
oversimplifies the issue by fixing on one single–and actually minor–causal
factor to the exclusion of many others. Accounts of such complex historical
episodes almost invariably oversimplify matters. It is inevitable for oversimple
thought about anything to be incomplete, because just this is exactly what
oversimplification is–the omission of significant detail through a failure to take

note of various factors that are germane to the matters at hand, thereby resulting
in a failure to understand the reality of things. Whenever we unwittingly
oversimplify matters we have a blind-spot where some significantly issue-
relevant facet of reality is concealed from our view.
Oversimplification involves a deficit of information. And this in turn
involves the incapacity to understand and explain phenomena. Thus if you
cannot tell the difference between A and a–if both of them look like  to you,
then the series
aAAXaaAaAAXaAaAAXAAXA . . .
will look like
XXXX. . .
to you. The X-occurrences seem now perfectly random. The fact that X
always and only follows the occurrence of a double A becomes lost.
And oversimplification can easily conceal structural information from our
sight. Thus consider simplifying I to II as per:
X X X X
I II
X X X X
One now loses sight of the fact that no X comes near the boundary as well
as gains the misimpression that all four quadrants have the same structure.
Optical myopia results involves indistinct vision, having an indefinite
image of things. Cognitive myopia is much like that, consisting in an imprecise
conception of things thanks to a lack of clarity in their conceptual (rather than
visual) apprehension. Both alike involve a failure to distinguish things that are
in fact distinct–or the reverse. There are two prime forms of cognitive myopia,
namely conflation when things that are different take to be alike (say both X and
Y seen as Z), and confusion when things that are the same take to be different
(say with one instance of X seen as such but the other as Y). Conflation often
has the result that irregularities are seen as regular; confusion has the result that
regularities are seen as irregular. Conflation often leads to seeing uniformity
where there should be difference. Confusion oversimplifies matters by seeing
difference where there is actually uniformity.
To see more clearly how oversimplification can create misunderstanding
consider someone whose visual myopia is such that he has is incompetent with
regard to telling 5 and 6 apart. As a result of such an inability to distinguish 5
from 6 the individual may well through conflation and failure to distinguish:
envision 56 as .
Or again, the individual may through confusion mix up 55 and 66 and so:
envision 56 as 65.
Conflation obscures sameness, confusion obscures difference. And neglect
of either results in oversimplification of complexity. Such modes forms of
cognitive myopia have very different ramifications for our grasp of the world’s
lawful comportment.
Suppose that we are in reality dealing with the perfectly regular series
R: 6 5 6 5 6 5 6 5 6 5 . . .
but due to the occasional confusion of a mild cognitive myopia we may then
actually „see” this (be it by way of observation or conceptualization) as
M: 6 5 5 5 6 5 5 5 6 5 . . .
But observe that our inability to distinguish has here effectively transmuted
a lawful regularity into a random disorder. It is then clear (via „Mill’s Methods of
Agreement and Difference”) that there is no causal correlation between R and M.
The supposition of (mild) myopia thus induces a drastic disconnection between
the two levels of consideration at issue, with the lawful order of R giving way to
lawlessness in regard to its model M.
Thus even so crude an example suffices to show that lawful order can
unravel and be destroyed by the confusion engendered by an occasional inability
to discern differences. And this relatively rudimentary observation has far-
reaching implications. In specific, it means that if even if the world is possessed
of a highly lawful order, this feature of reality may well fail to be captured in even
a mildly myopic representation of it.
Again suppose realty as per I is oversimplified as per II:
X
X
X
I II
X
X
The result is clearly an invitation to mistakes. The absence of diagonal

symmetry is lost to our sight. And the difference between the X-containing
quadrants also becomes unavoidable. We are bound seriously to mis-appreciate
the nomic structure of the situation.
One of the salient aspects of oversimplification lies in the fundamental
epistemological fact that errors of omission often carry errors of commission in
their wake: that ignorance plunges us into actual mistakes.
Where Reality is concerned, incompleteness in information invites

incorrectness. If you do not know the missing letter in diagram A you are likely
to think it too as an X and so take the situation to be as per B instead of C.
(A) (B) ©)
X X X X X X X X X
X X X X o X
X X
X X X X X X X X X
Your ignorance then engenders the misleading impression that the rows are
all alike instead of realizing the differences at issue.
In oversimplifying we slide into falsification. Is oversimplification thereby
lying? No! And for good reason.
Lying is falsification with the intent to deceive. Falsity is here the aim of the
enterprise. But oversimplification is falsification with the intent to convey truth.
Falsity is here merely collateral damage.
And there is a further crucial difference. The liar affirms what he believes
to the false: to lie is to affirm a recognized falsehood. In oversimplification
we do indeed commit to falsehoods: but we do not realize what they are. The
oversimplifier is agnostic about the falsity of which he commits. Lying in
consequence is ethically reprehensible, while oversimplification is in general
ethically venial because inadvertant.
3. Conclusion
It is time to conclude. The jist of the present discussion can be summarized
in a few lines effectively consisting in the following Q/A sequence:
– Why do we simplify matters? A: Because this provides us with
indispensable economies of procedure.
– Why do we oversimplify? A: Because the difference between
constructive simplification and oversimplification is something that
can be discerned only ex post facto with the wisdom of hindsight.
– What does oversimplification portend? A: Oversimplification always
carries the risk of error and misunderstanding.
– Why risk the negativities of oversimplification? A: Because there is no help
for it; even in natural science it is an unavoidable part of rational procedure.
It has been the aim of the project to expound and explain this line of
thought–and to exhibit in detail the consequences and implications of the
unwelcome realties that ensue in its wake.4
4 Paper presented in the Luncheon Lecture series at the University of Pittsburgh’s Philosophy
of Science Center in September of 2014.
Werner Sauer Original Scientific Paper
University of Graz UDK 111 Аристотел
14 Аристотел
Metaphysics EZHΘ: Aristotle’s Unified Treatise περὶ τῆς

οὐσίας καὶ τοῦ ὄντος, de Essentia et de Ente
I
First, let us state the issue.1 The Metaphysics text in its present form consisting
of fourteen books is evidently, „on the face of it”, as Ross in the Introduction to
his great Metaphysics edition puts it, „not a single finished work”.2 This much
is common consent, for since from Werner Jaeger onwards the Chorizontes
(οἱ χωρίζοντες), as we may call them by an old name originally given to those
ancient grammarians who split up Homer into two, scrutinized the structure of
the work, its unity vanished once and for all; these efforts at dissecting it into
strata and treatises were carried to the extreme by Ingemar During, who viewed
the Metaphysics as nothing more than a collection of separate treatises, calling
them Schrift Alpha, Schrift A elatton, Schrift Beta, and so on.3
But even During left untouched the unity, evident as it is, among the
three books ZHΘ – Jaeger’s Substanzbücher –, calling them the Schrift ZHΘ on
sensible substances, περὶ οὐσιῶν αἰσθητῶν.4 On the other hand, it is common
wisdom that the short book E does not belong to the treatise ZHΘ but (During’s
radicalism set aside) rather to the preceding book Γ, as the end of a treatise
ΑΒΓΕ, as is argued, e.g., by Ross.5
Now, this tearing off Ε from the following book Z is bound to strike the
innocent reader of the Metaphysics as odd, and even more so when finding the
very same Ross stating this apparent plain truth that „[t]he substantial continuity
of ΖΗΘ with E is [...] evident”:6 how can one and the same Aristotle scholar, and
a great one indeed, both affirm a substantial continuity between the two books
1 This paper is based on an honorary doctorate address at the University of Belgrade.

2 Ross W. D., Aristotle’s Metaphysics. A Revised Text with Introduction and Commentary (1924),
2 vols, reprint with corrections Oxford: Clarendon Press, 1953, vol. I, p.xiii.
3 Ingemar During, Aristoteles. Darstellung und Interpretation seines Denkens, Heidelberg:
Winter, 1966, p.586–93.
4 Ibid. p.589, 593.
5 Ross [see n.1], vol. I, p.xviif.; see also Michael Frede & Gunther Patzig, Aristoteles ‚Metaphysik‘.
Text, Ubersetzung und Kommentar, 2 vols, Munich: Beck, 1988, vol. I, p.27f. – The books α
and Δ (Aristotle’s ‘philosophical dictionary’) are commonly agreed to be later additions to the
Metaphysics, and are in any case not our concern.
6 Ross [see n.1], vol. I, p.xxi.
Ε and Z and yet deny that they are parts of what originally was a single unified
treatise?
Of course, that substantial continuity Ross is speaking of was earlier
noted as well by those luminaries of 19th century Aristotle scholarship, Bonitz
and Schwegler. Bonitz, although believing the seven books ΑΒΓΕΖΗΘ to have
been written uno tenore et eodem disputationis contextu, nevertheless accorded
a special unity to the books ΕΖΗΘ which he says ita inter se sunt connexi, ut
alterum divellere ab altero aegre sustineas, and, secondly, that in E, as if it were
the beginning of a new treatise altogether, quasi de integro disputationem videtur
ordiri;7 and Schwegler even held the severing of Ε from Z as a separate book to
be a blunder of some early editor of the Metaphysics.8
By and large, we will in the following defend the view of Bonitz and of
Schwegler to the extent that Ε belongs together with ΖΗΘ, though conceding
to our more recent Chorizontes a split, albeit of course not one between E and
Z, but between E and the preceding book Γ, which was seen clearly by Bonitz
already (‘quasi de integro etc.’) who, however, shunned back from drawing the
natural conclusion (why so, will come forth in sec. III); and we shall also see that
to some extent at least Schwegler’s contention that E and Z are even one single
book is perfectly defensible.
II
Perhaps one would now expect the substantial connection of book E with
book Z which we have made so much of to be fleshed out in more detail. But
at this point this would be premature, and in any case it will become gradually
apparent as we proceed. Presently, let us simply look at certain textual evidences
for the unity among the Substanzbücher which are usually taken as well as
evidences for them having originally been a separate treatise.
These are, first of all, three back references, two of them in book Θ. The first
of these two is the beginning of the book:
[1] Now we have treated of what primarily is being and to which all the
other categories of being are referred, i.e., substance (περὶ μὲν οὖν τοῦ πρώτως
ὄντος καὶ πρὸς ὃ πᾶσαι αἱ ἄλλαι κατηγορίαι τοῦ ὄντος ἀναφέρονται εἴρηται, περὶ
τῆς οὐσίας).
For it is in virtue of the formula of substance that the others are called
beings, i.e. quantity and quality and the like; for they all will contain the formula
of substance, as we said in the first discussions (κατὰ γὰρ τὸν τῆς οὐσίας λόγον
λέγεται τἆλλα ὄντα, τό τε ποσὸν καὶ τὸ ποιὸν καὶ τἆλλα τὰ οὕτω λεγόμενα·
7 Hermann Bonitz, Commentarius in Aristotelis Metaphysicam (1849), reprint Hildesheim:

Olms, 1992, p.10, 14.
8 “[D]as sechste Buch hängt so unmittelbar [...] mit dem siebenten zusammen, das man beide
nur als Ein Buch auffassen, und ihre ganz sinnlose Zerreisung nur für ein Misverständnis
des Ordners der Metaphysik ansehen kann”. Albert Schwegler, Die Metaphysik des Aristoteles.
Grundtext, Übersetzung und Commentar nebst erlauternden Abhandlungen (1847–48), 4 vols,
reprint Frankfurt am Main: Minerva, 1960, vol. IV, p.1.
Werner Sauer : Metaphysics EZHΘ: Aristotle’s Unifed Treatise περὶ τῆς οὐσίας καὶ τοῦ ὄντος... 95
πάντα γὰρ ἕξει τὸν τῆς οὐσίας λόγον, ὥσπερ εἴπομεν ἐν τοῖς πρώτοις λόγοις)
(Θ.1 1045b27–32).9
The reference is to Z.1, where Aristotle argues that all the other categorial
beings, the accidents, are beings only in virtue of being quantities, qualities etc. of
substance,10 and that therefore substance is primary in every respect, including
primacy ‘in formula’ (τῷ λόγῳ 1028a32, 34), that is, in account or definition.11
We have rendered the phrase ἐν τοῖς πρώτοις λόγοις in passage [1] as ‘in
the first discussions’, these πρῶτοι λόγοι being of course about primary being
i.e. substance mentioned right before. Now the word λόγοι when used for
referring to some other text may of course also mean ‘treatise’, or ‘tract’, but here
this cannot be the case, since ‘in the first treatise’ would make no sense: thus,
Aristotle refers back to Z.1 as the beginning of his discussion of substance in ZH
within a work to which Θ itself belongs.12
The other of those two back references is Θ.8 1049b27–29:
[2] It has been said in our discussion of substance that everything which
comes to be, comes to be something from something and by something, and this
is the same in species as it (εἴρηται δὲ ἐν τοῖς περὶ τῆς οὐσίας λόγοις ὅτι πᾶν τὸ
γιγνόμενον γίγνεται ἔκ τινος τὶ καὶ ὑπό τινος, καὶ τοῦτο τῷ εἴδει τὸ αὐτό).
The reference is to Z.7 1032a13f., a24f., Z.8 1033a24–27, b29–32; qua
reference, this one differs from the former only by saying explicitly that the
earlier discussion it refers to is about substance, περὶ τῆς οὐσίας: but this is only
to be expected because unlike the former reference it does not have mentioned
the topic of those λόγοι right before in the preceding sentence.
Both Ross and Frede & Patzig claim that the Θ references indicate that
originally book Z was the beginning of a separate treatise distinct from the
books ΑΒΓΕ. Ross says „ZHΘ evidently form a fairly continuous work”, which of
course is uncontested, but to add:
9 The translation, as in the following Metaphysics quotes, is based on, but by no means identical
with, Ross’s, in vol. II of The Complete Works of Aristotle. The Revised Oxford Translation, 2
vols. Ed. by Jonathan Barnes, Princeton: Princeton UP, 1984. The Greek Metaphysics text we
use is Jaeger’s edition, W. Jaeger ed., Aristotelis Metaphysica, Oxford: Clarendon Press, 1957.
Citations from other works of Aristotle are taken from the Bekker edition, Aristotelis Opera,
2 voll. Ex rec. Immanuelis Bekkeri. Ed. altera quam curavit Olof Gigon, Berlin: de Gruyter,
1960. (Our numbering of Aristotle quotes is intended to facilitate cross references).
10 They are not entia properly speaking but only entis, as Aquinas puts it, or in Alexander’s
phrase, they are only οὐσίας τι, something-of-substance. See S. Thomae Aquinatis in
duodecim libros Metaphysicorum Aristotelis Expositio. Ed. M.-R. Cathala & R. M. Spiazzi,
Turin; Rome: Marietti, 1964, lib. XI, lect. 3, §2197; lib. XII, lect. 1, §2419. For Alexander’s
phrase, see Alexandri Aphrodisiensis in Aristotelis Metaphysica Commentaria. Ed. Michael
Hayduck, Berlin: Reimer, 1891 (CIAG I), p.242, ll.22, 23 (fuller quote in n.24 below). – As to
Aristotle himself, see the Z.1 part of citation [9] below.
11 “Necessarily in the formula of each [of the other] things the formula of substance is
contained”, ἀνάγκη γὰρ ἐν τῷ ἑκάστου λόγῳ τὸν τῆς οὐσίας ἐνυπάρχειν Z.1 1028a35f.
12 That is, Θ.1–9, dealing with potential and actual being; Θ.10, turning to the topic of being as
the true (and non-being as the false), is a different matter altogether.
It is evident, again, that the reference of Θ to Z as ‘our first discussions’

implies that ZHΘ is in a sense a distinct treatise from ΑΒΓΕ (Ross [see n.1], vol.
I, p.xviii).
To this, Frede & Patzig only add that the Θ references refer to ZH as an originally
separate treatise on substance, to which Aristotele afterwards attached Θ.13
Thus, both Ross and Frede & Patzig take it for granted that the πρῶτοι
λόγοι of passage [1] are first discussions – on substance of course – in a separate
treatise: but the addition ‘in a separate treatise’ is wholly unwarranted, as we have
seen; and if so, the reference in [2] to λόγοι περὶ τῆς οὐσίας does not indicate
anything in their support either.
In arguing for their view, Frede & Patzig (see quote in n.12) draw attention
to yet another back reference, which is in /.2, referring back to the central thesis
of Z.13:
[3] Now if no universal can be substance, as has been said in the logoi on
substance and being (εἰ δὴ μηδὲν τῶν καθόλου δυνατὸν οὐσίαν εἶναι, καθάπερ ἐν
τοῖς περὶ οὐσίας καὶ περὶ τοῦ ὄντος εἴρηται λόγοις) [...] (1053b17f.).
We have left λόγοις untranslated in order not to prejudge the issue: does
the word here simply mean ‘discussions’ leaving open whether in the same work
or in another, or is it to be understood as ‘treatise’? Well, in view of the fact that
no-one will take book / as having originally belonged together with the books
ZHΘ,14 here the word actually refers to a treatise on its own. This much, then,
Frede & Patzig are perfectly right. But they simply ignore that Aristotle does
not call that treatise as being on substance, περὶ οὐσίας, only, but as being on
substance and on being, περὶ οὐσίας καὶ περὶ τοῦ ὄντος, with οὐσία literally
translated into Latin, de essentia et de ente: and to have anything to which this
περὶ τοῦ ὄντος could be geared, we must look beyond Z, which already is περὶ
οὐσίας, to the preceding book Ε, stating in its first sentence that the incipient
work is one about beings, or things-that-are, in general:
[4] We are seeking the principles and the causes of the things-that-are, and
plainly of them qua things-that-are (αἱ ἀρχαὶ καὶ τὰ αἴτια ζητεῖται τῶν ὄντων,
δῆλον δὲ ὅτι ᾗ ὄντα) (E.1 1025b3f.), which is to say that this inquiry will be περὶ
ὄντος ἁπλῶς (b9), about being simply.
13 “Man wird also sagen durfen, das Z und H eine ursprungliche Einheit bildeten, das es
aber fraglich ist, ob Θ schon von Anfang an Teil dieser Einheit war. Nennen wir diese
ursprungliche Einheit, Θ 8 folgend, die Ahandlung „Über die ousia“.
Manches spricht dafür, das es sich bei dieser Abhandlung „Über die ousia“ um eine
ursprunglich selbständige Abhandlung handelte, die erst später, nun schon durch Θ 1–9
ergänzt, als Kernstück für die „Metaphysik“ verwendet wurde. Nicht nur bezieht sich Θ 8,
1049b27–28, auf ZH als die Abhandlung über die ousia, auch I 2, 1053b17–18, verweist auf
ZH als „die Ausfuhrungen uber die ousia und das Seiende“ [...], so als handle es sich um
eine selbständige Schrift. Z beginnt ohne Verbindungspartikel, so als handle es sich um den
Anfang einer ganzen Schrift. Dazu past auch, das Θ 1, 1045b32, sich auf den Anfang von Z
so bezieht, als handle es sich um den Anfang einer Abhandlung”. Frede & Patzig [see n.4],
vol. I, p.22f.; cf. p.28f.).
14 As Ross [see n.1], vol. I, p.xxii puts it, “I is evidently a more or less self-contained treatise,
dealing with the nature of unity and of kindred conceptions”.
Hence we cannot but conclude that the reference in I.2 is to a treatise

beginning with book E, treating of being in general, and continuing with the
Substanzbücher focussing on what is primary being, such that E will not belong
to the preceding books ΑΒΓ but will be the beginning of a unified treatise EZHΘ
de ente et essentia, covering the ‘de ente’ part.
Thus the three back references in Θ.1, Θ.8, and in I.2 respectively, do nothing
to support the contention of our Chorizontes, while the last one rather supports
the opposite view, viz., that E belongs together with ZH0.
However, they adduce still another textual phenomenon in support of their
view: namely the fact that there are connecting particles at the beginning of H
and Θ (δή and μέν, resp.) but none at the beginning of Z, this book beginning
baldly with τὸ ὂν λέγεται πολλαχῶς 1028a10, so as if it were the beginning of a
separate treatise. 15
But now, look back at passage [4]: it is the first sentence of E, but has no
connecting particle either, such that by parity of reason one should rather infer
that E is the beginning of a separate treatise but not a continuation of Γ. Indeed,
in the latter case one would expect an even stronger connecting device. In that
first sentence of E Aristotle takes up the conclusion of Γ.1 (quote in n.18 below),
such that on the presupposition of an original connection of E with Γ one would
expect some referring remark such as ώσπερ εϊρηται πρότερον, ‘as has been said
earlier’, or the like; but there is none, and we are reminded of Bonitz’ ‘quasi de
integro etc.’, only that by now we are strongly inclined already to drop its ‘quasi’.
In a word, the lack of a connecting particle at the beginning of Z proves
nothing, and moreover, we will see later on that this lack can anyway quite easily
be treated as a mere pseudo-obstacle.
So far, then, the claim that Z was originally the beginning of a separate
treatise to which E did not belong has only very feeble support. But there are also
stronger reasons behind that claim, and they are to be found in book K.
III
Book K is a somewhat odd thing. It consists of two clearly distinguishable
parts, of which the first, K.1–8, is a parallel version of the content of the books
BΓE, whereas the second, K.9–12, consists of almost literal excerpts from the
Physics16 and hence does not concern us here at all.
Now from the fact that K.1–8 corresponds to BΓE, and from the general
assumption that it is a very early text, the conclusion is ordinarily drawn that E
actually is connected with B and Γ, and not with the books ZHΘ which certainly
were written fairly late. Ross, for instance, says:
The doubt which has sometimes been expressed on the question whether
there is any real connexion between E and ΑΒΓ is set at rest by the fact that the
first part of K [K.1–8], which is certainly very old and may well be a pupil’s notes
15 See Ross [see n.1], vol. I, p.xviii, and for Frede & Patzig the quote in n.12.
16 More precisely, the dividing line is in K.8 1065a26.
of a course of lectures by Aristotle himself, is a continuous parallel treatment of

the topics discussed in ΒΓΕ.17
The same view is expressed by Frede & Patzig.18 And to add, it was nothing
else but K.1–8 that hindered Bonitz to draw from his insight that in E the
discussion starts anew, the conclusion that E cannot be viewed as a continuation
of the discussion in Γ and may, as the beginning of a new treatise EZHΘ, even
have been intended by Aristotle himself to be completely separated, omnino
seiungi, from ABΓ: but any such query, Bonitz ([see n.6], p.14) says, tolli videbitur
comparato libro K. Now to see what this K spell, as it may be dubbed, is all about,
we have to cast a look at K ‘s chapter 3, corresponding to Γ.1–2, and at its chapter
8, corresponding to E.2–4.
As to K.3, it will be fitting first to look at Γ.1–2. In Γ.1, Aristotle posits a
science of being qua being, and argues to the conclusion that this science
investigates the first principles and causes of being qua being.19 Then, Γ.2 starts
with stating that
[5] being is spoken of in many ways, but in relationship to one single thing,
one definite nature, and not homonymously ((τὸ δὲ ὂν λέγεται μὲν πολλαχῶς,
ἀλλὰ πρὸς ἓν καὶ μίαν τινὰ φύσιν καὶ οὐχ ὁμωνύμως) (Γ.2 1003a33f.)
which is to say that although ‘being’, όν, is not a genus word such that what-
is does not exhibit the unity of a genus,20 it nevertheless does not bear the same
name ‘being’ equivocally (ὁμωνύμως), by mere linguistic accident21 – just as
little as what is called healthy (the body, the food, the climate etc.) does –, but
everything (ἅπαν 1003b5) called a being is so called – again as it is in the case
of what is called healthy – in virtue of a πρὸς-ἕν or focal-meaning relationship
„with respect to one single principle (πρὸς μίαν ἀρχήν)” (b6):
[6] For some things are called beings because they are substances (τὰ μὲν
γὰρ ὅτι οὐσίαι, ὄντα λέγεται),
others, because they are affections of substance (τὰ δ᾽ ὅτι πάθη οὐσίας),
and still others because they are a way towards substance (τὰ δ᾽ ὅτι ὁδὸς εἰς
οὐσίαν)
17 Ross [see n.1], vol. I, p.xviif. He even finds it apt, ibid. p.xxvi, to „even conjecture that K
represents an earlier course than BΓE“. This was definitely asserted by Jaeger: see Werner
Jaeger, Aristoteles. Grundlegung einer Geschichte seiner Entwicklung, Berlin: Weidmannsche
Buchhandlung, 1923, p.216ff. See also end of sec. V.
18 Frede & Patzig [see n.4], vol. I, p.28: “Das E zu der Reihe A, B, Γ gehort, wird [...] auch
dadurch belegt, das K in seinem ersten Teil den Gedankengang von BΓE kontinuierlich in
einer Parallelfassung vortragt (K 1–8 ist vermutlich die Nachschrift eines Schulers von einer
Vorlesungsreihe, die Aristoteles selbst einmal vorgetragen hat)“.
19 “There is a science which investigates being qua being and those things which belong to this
in virtue of itself “, ἔστιν ἐπιστήμη τις ἣ θεωρεῖ τὸ ὂν ᾗ ὂν καὶ τὰ τούτῳ ὑπάρχοντα καθ᾽
αὑτό 1003a21f. – „Therefore we also must grasp the first causes [and principles] of being qua
being“, διὸ καὶ ἡμῖν τοῦ ὄντος ᾗ ὂν τὰς πρώτας αἰτίας [καὶ ἀρχὰς a29] ληπτέον a31f.
20 οὐ γὰρ γένος τὸ ὄν, An. Post. II.7 92b14.
21 As, e.g., both the donkey and the engine are called ὄνος. The example is in Topics I.15
107a19f.
or destructions or privations or qualities or productive or generative of

substance or of things spoken of with reference to substance (ἢ φθοραὶ ἢ στερήσεις
ἢ ποιότητες ἢ ποιητικὰ ἢ γεννητκὰ οὐσίας ἢ τῶν πρὸς τὴν οὐσίαν λεγομένων),
or negations of an instance of these or of substance (ἢ τούτων τινὸς
ἀποφάσεις ἢ οὐσίας) (1003b6–10).
Of this list, we note here just this one feature, which will show to be
important, namely, that it covers among the beings spoken of with reference
to substance not only things falling into the non-substantial categories, as is
most obvious in the case of the qualities, but also things outside the categories
altogether, as again is most obvious in the case of privations (e.g., blindness) and
negations (e.g., not-man, not-white and the like, cf. especially N.2 1089a16–19).
Then, in 1003b11–19, Γ.2 goes on to argue that for the unity of a science it
is sufficient that its domain has the unity of a focal-meaning structure, and that
a science with such a domain studies everything in it:
[7] Just as there is one science which deals with all the healthy things, this
is equally so in the other cases (καθάπερ οὖν καὶ τῶν ὑγιεινῶν ἁπάντων μία
ἐπιστήμη ἔστιν, ὁμοίως τοῦτο καὶ ἐπὶ τῶν ἄλλων) b11f.,
but it studies chiefly its primary item (κυρίως τοῦ πρώτου ἡ ἐπιστήμη b16),
and hence it is plain that
(i) it belongs to a single science to study all the things-that-are qua things-that-
are (δῆλον οὖν ὅτι καὶ τὰ ὄντα μιᾶς [sc. ἐπιστήμης] θεωρῆσαι ᾗ ὄντα b15f.),
and, since substance is the primary item in this all-encompassing domain, that
(ii) the philosopher must grasp the principles and causes of substances (τῶν
οὐσιῶν ἂν δέοι τὰς ἀρχὰς καὶ τὰς αἰτίας ἔχειν τὸν φιλόσοφον b18f.).
This much is from Γ.2 enough for our purposes. Let us now turn to K.3.
K.3, from its beginning 1060b31 up to 1061a10 corresponds to Γ.1 and to Γ.2
till the end of the list of being at 1003b10. It will be sufficient to quote only the
corresponding K.3 list. Being is a πολλαχῶς λεγόμενον (1060b32f.), just as for
instance the healthy things are, but just as these display a focal meaning structure,
[8] in the same way everything that-is is spoken of (τὸν αὐτὸν δὴ τρόπον
καὶ τὸ ὂν ἅπαν λέγεται): for in virtue of being an affection or a permanent or
a transient state or a movement of being qua being, or something else of this
sort, each of these things is called a being (ττῷ γὰρ τοῦ ὄντος ᾗ ὂν πάθος ἢ ἕξις
ἢ διάθεσις ἢ κίνησις ἢ τῶν ἄλλων τι τοιούτων εἶναι λέγεται ἕκαστον αὐτῶν ὄν)
(K.3 1061a7–10),
such that by this focal meaning structure „every being is referred to some one
common thing”, παντὸς τοῦ ὄντος πρὸς ἕν τι καὶ κοινὸν ἡ ἀναγωγὴ γίγνεται a10f.
Somewhat odd here is the use of ὂν ᾗ ὄν, since here it must mean
‘substance’.22 As to πάθος, ἕξις and διάθεσις, we best take these words to refer
22 Thus, Ps.-Alexander in his paraphrase of 1061a7–10 replaces ὂν ᾗ ὄν without any comment

by οὐσία: “[...] in the same way [as in the case of what is healthy] the things-that-are too
to the affective qualities, the third class, and to the more stable and less stable
states, e.g. knowledge versus health, these being the first class in the category of
qualities in Categories chapter 8; and κίνησις certainly stands, as it often does, for
the categories of acting, ποιεῖν, and being acted upon, πάσχειν.23
But then, the main difference between this K.3 list [8] and the Γ.2 list [6]
is that the K.3 list is confined to categorial being only, expressly called τὸ ὂν
ἅπαν, the entirety of being, apart from substance consisting of „quantity, quality,
where, and the other categories”, as Ps.-Alexander, commenting on 1061a10f.,
puts it.24
And this difference is all-important. For on the Γ.2 approach by way of the
πρὸς-ἕν structure of being there is no basis nor any motivation to separate out
from the entirety of what is called being first categorial being, and to set aside
other sorts of being such as privations and negations:
(a) no basis, since the fundamental dividing line is the one between being
strictly so called, i.e. substance, and what is, in Alexander’s phrase which
occurs precisely in his commentary on [6], only οὐσίας τι, something-of-
substance, and of such kind is a privation just as well as a non-substantial
categorial being, a quality, a quantity, and the like;25 and
(b) no motivation rergarding the determination of the proper object of
the science of being qua being, since according to the Γ.2 approach a
science whose domain forms a πρὸς-ἕν structure studies everything
within it (passage [7]), albeit primarily its ἕν, that is substance in the
case of this particular science; thus, even non-categorial beings such
as privations or negations will not be wholly excluded from it. And to
add, it is telling that in K.3 after 1061a10f. nothing follows which would
correspond to Γ.2 1003b11–19, where Aristotle argues for this.
Thus, the K.3 list of being, already confined as it is to categorial being,
presupposes some eliminating procedure, which cannot stem from the notion of
have got this name from some one thing, viz., substance, which is being in the principal
sense”, οὕτω καὶ τὰ ὄντα ἀπὸ ἑνός τινος, τῆς οὐσίας, ἥπερ κυρίως ὄν ἐστι, τὴν προσηγορίαν
ταύτην εἴληφε. Alexandri etc. [see n.9], p.642 ll.22f. Similarly Aquinas ([see n.9], lib. XI, lect.
3, §2197) identifies the ens inquantum ens of his Latin Metaphysics with ens simpliciter, which
is what in se habet esse, scilicet substantia.
23 See, e.g., Ethica Eudemia I.8 1217b29, or Z.4 1029b25, and Frede & Patzig [see n.4] ad loc.,
vol. II, p.62f.
24 εἰσὶ δὲ ταῦτα, i.e. the things referred to some one common thing, ποσὸν ποιὸν ποῦ καὶ αἱ
λοιπαὶ κατηγορίαι.
Alexandri etc. [see n.9], p.642 ll.29f. The same confinement to categorial being is taken for
granted in Aquinas’ comments on the K.3 list, see the reference in n.21.
25 In fact, Alexander applies the phrase in particular to non-categorial items, among them
privations, presumably in order to emphasize their status as beings: “And both the destructions
and changes of substance we call beings, because these too [and not only such things as the
πάθη which are qualities] are something-ofsubstance [...] And likewise the privations too: for
blindness is a being because it is something-of-substance”, καὶ τὰς φθορὰς δὲ τῆς οὐσίας καὶ
μεταβολὰς ὄντα λέγομεν, ὅτι καὶ ταῦτα οὐσίας τί ἐστιν […] ὁμοίως καὶ αἱ στερήεις· ἡ γὰρ
τυφλότης ὄν τι ὅτι οὐσίας τι. Alexandri etc. [see n.9], p.242 ll.21–23.
a πρὸς-ἕν structure, which is to say that the K.3 list cannot be understood with
reference to Γ.2 only.
On the other hand, it bears a striking resemblance to something in Z.1.
After having stated at the beginning of Z.1 once more that being, here meaning
categorial being, λέγεται πολλαχῶς 1028a10, and having argued that substance
is the primary being, he continues with what we best present face to face with
the K.3 list:
K.3: in the same way as the healthy Z.1: primary being is substance,
[9] everything that-is is spoken of (τὸ ὂν ἅπαν λέγεται): the other things [the accidents]
are called
for in virtue of being an affection or a permanent beings (τὰ δ᾽ ἄλλα λέγεται ὄντα)
or a transient state or a movement of being qua in virtue of some of them being
quantities of
being, or something else of this sort being in that sense, others
qualities of it,
(τῷ γὰρ τοῦ ὄντος ᾗ ὃν πάθος ἢ ἕξις ἢ διάθεσις others affections of it, and others
something
ἢ κίνησις ἢ τῶν ἄλλων τι τοιούτων εἶναι), else of this sort (τῷ τοῦ οὕτως
ὄντος τὰ μὲν
each of these things is called a being ποσότητες εἶναι, τὰ δὲ
ποιότητες, τὰ δὲ πάθη,,
((λέγεται ἕκαστον αὐτῶν ὄν) (1061a8–10). τὰ δὲ ἄλλο τι τοιοῦτον)
(1028a18–20).
The two passages resemble each other, both in expression by their τῷ …
ὄντος … εἶναι and materially by their taking into account categorial being only,
to such an extent that one cannot help but to surmise that the K.3 list of being
relates really rather to Z.1 as its background, and not to Γ.2.
But in order to see more clearly what is going on here, and to pin down the
eliminating procedure that contracts the indiscriminate Γ.2 list of being to what
we have in K.3, we must now turn to E.2–4 to which in Κ corresponds chapter 8.
IV
Having in E.1 re-opened the quest for the principles and causes of being
qua being (passage [4]), and having towards the end of it identified the science
of supreme being (ἡ θεολογικὴ ἐπιστήμη) with the science of being qua being
i.e. of being in general – more on this will come in sec. V, (γ) –, Aristotle at the
beginning of E.2 once again says, as he had done at the beginning of Γ.2 (passage
[5]), that being is a πολλαχῶς λεγόμενον. But the way he now explains this is
entirely different from what we have in Γ.2, by taking up the fourfold partition
of being in Δ.7:
[10] But since being, so called without qualification, is spoken of in many
ways (ἀλλ᾽ ἐπεὶ τὸ ὂν τὸ ἁπλῶς λεγόμενον λέγεται πολλαχῶς), of which
[a] one was [in Δ.7] accidental being (ὧν ἓν μὲν ἦν τὸ κατὰ συμβεβηκός [sc. ὄν]),
[b] another being as the true, and non-being as the false (ἕτερον δὲ τὸ ὡς
ἀληθές, καὶ τὸ μὴ ὂν τὸ ὡς ψεῦδος),
[c] but apart from these there are the figures of predication (παρὰ ταῦτα δ᾽
ἐστὶ τὰ σχήματα τῆς κατηγορίας), e.g., what [a thing is, i.e., substance], of what
quality, of what quantity, where, when, and anything else that signifies in this way
(οἷον τὸ μὲν τί, τὸ δὲ ποιόν, τὸ δὲ ποσόν, τὸ δὲ πού, τὸ δὲ ποτέ, καὶ εἴ
τι ἄλλο σημαίνει τὸν τρόπον τοῦτον),
[d] again apart from all these, potential and actual being (ἔτι παρὰ ταῦτα
πάντα τὸ δυνάμει καὶ ἐνεργείᾳ [ὄν]) [...] (E.2 1026a33-b2).
No word of a πρὸς-ἕν, or focal-meaning structure here. Let us further note
just this point, that although the third member of this list, which is categorial
being, is itself again a πολλαχῶς λεγόμενον, in that the categories are summa
genera (see, e.g., Δ.28 1024b12–16), this is not mentioned here in E.2 at all,
whereas it is mentioned in Δ.7:
[11] For ‘to be’ signifies in as many ways as the figures of predication are
spoken of ([...] τὰ σχήματα τῆς κατηγορίας· ὁσαχῶς γὰρ λέγεται, τοσαυταχῶς τὸ
εἶναι σημαίνει) (1017a23f.);
and this again is something which will prove significant.
After introducing the fourfold partition of being, Aristotle in E.2 immediately,
with the words
[12] Now since being is spoken of in many ways, we must first say concerning
accidental being that there is no scientific study of it (ἐπεὶ δὴ πολλαχῶς λέγεται
τὸ ὄν, πρῶτον περὶ τοῦ κατὰ συμβεβηκὸς λεκτέον, ὅτι οὐδεμία ἐστὶ περὶ αὐτὸ
θεωρία) (1026b2–4),
turns to the discussion, lasting till the end of chapter 3, of accidental being,
that is, beings that are compounds resting on accidental predications, e.g.,
musical Socrates who exists because Socrates is musical; and then in chapter 4
he discusses more briefly being as the true which rests on the use of ἔστι in
the sense of ‘is true’ and consists in such judgment-contents as, e.g., Socrates is
musical, or Socrates is not pale.26
And both these modes of being are excluded from the domain of the science
of being qua being (1027b28–1028a3): accidental being has no determinate
cause such that it cannot be studied by any science, and being as the true is
disqualified by its being mind-dependent, being „an affection of thought”, τῆς
διανοίας τι πάθος 1027b34–28a1. There is, however, also a common reason for
their exclusion, which is much more interesting:
[13] and both are around the remaining genus of being, and do not indicate
there being any extra nature of being (καὶ ἀμφότερα περὶ τὸ λοιπὸν γένος τοῦ
ὄντος, καὶ οὐκ ἔξω δηλοῦσιν οὖσάν τινα φύσιν τοῦ ὄντος) (1028a1f.),
that is to say that they are parasitical upon categorial being: musical Socrates
does indicate the categories of substance and of quality, but ‘no extra nature of
being’, and the same holds of the being as the true Socrates is musical.
26 An example for this alethic use of ἔστι by Aristotle himself is in the beginning of the Posterior
Analytics: To acquire knowledge we must beforehand assume, for instance, „that in every
case either to affirm or to deny it is true (ὅτι μὲν ἅπαν ἢ φῆσαι ἢ ἀποφῆσαι ἀληθές), that it is
(ὅτι ἔστι)“, An. Post. I.1 71a13f.
Thus, „these two are to be dismissed”, ταῦτα μὲν ἀφείσθω 1028a3: „But what
is to be investigated”, σκεπτέον 1028a3, „are the causes and principles of being
itself qua being”, τοῦ ὄντος αὐτοῦ τὰ αἴτια καὶ τὰς ἀρχὰς ᾗ ὄν a3f.
From the context it is plain that this being itself, a few lines before, 1027b31,
called κυρίως ὄν, principal being, with substance, quality, and quantity as
examples, is categorial being, being ‘according to the figures of predication’.
And with that we finally have the eliminating procedure we need to
understand the restriction of the Γ.2 list of being to categorial being in the
corresponding K.3 list: in a word, the K.3 list presupposes E.2–4, or within K,
chapter 8. Let us now look briefly at this chapter.
K .8 does not start, as E.2 does (passage [10]), with presenting first the
fourfold partition of being according to Δ.7, but presupposes it as familiar:
though it begins with almost the same words as E.2 up to the mention of
accidental being, it then proceeds, without any enumeration of the other three
modes of being, straightforwardly to the discussion of that mode:
[14] But since being, without qualification, is spoken of in several ways (ἐπεὶ
δὲ τὸ ἁπλῶς ὂν κατὰ πλείους λέγεται τρόπους), of which one is what is called
to be accidentally (ὧν εἷς ἐστὶν ὁ κατὰ συμβεβηκὸς εἶναι λεγόμενος), we must
consider first what is being in this way (σκεπτέον πρῶτον περὶ τοῦ οὕτως ὄντος)
(K.8 1064b15–17).
Even more obvious becomes that presuppositon when within the discussion
of accidental being as contrast to it categorial being shows up in passing, which
is here, as in Δ.7 but not in E.2 where Aristotle speaks only of ‘the figures of
predication’, called ὂν καθ᾽ αὑτό, being in its own right, or in virtue of itself:27
[15] But it is plain that of accidental being (τοῦ κατὰ συμβεβηκὸς ὄντος)
there are no causes and principles of the same kind as those of being in its own
right (τοῦ καθ᾽ αὑτὸ ὄντος) (K.8 1065a6–8).
And similarly out of the blue, so to speak, being as the true shows up in the
closing lines of the E.2–4 part of K.8 (cf. n.15), corresponding to E.4 1027b29–
1028a3 where being as the true, together with coincidental being, is dismissed:
[16] But as to being as the true and accidental being (τὸ δ᾽ ὡς ἀληθὲς ὂν
καὶ τὸ κατὰ συμβεβηκός), the former lies in a combination of thought and is
an affection of it – therefore it is not the principles of being in this sense which
are sought, but of being which is outside and separable (περὶ δὲ τὸ ἔξω ὂν καὶ
χωριστόν). And the latter, I mean accidental being (λέγω δὲ τὸ κατὰ συμβεβηκός).
And the latter, I mean accidental being (λέγω δέ τό κατά συμβεβηκός), is not
necessary but indetermined, and its causes are unordered and indefinite (K.8
1065a21–26).28
27 Δ.7 1017a7f.: τὸ ὂν λέγεται τὸ μὲν κατὰ συμβεβηκὸς τὸ δὲ καθ᾽ αὑτό – a22f.: καθ᾽ αὑτὰ δὲ
εἶναι λέγεται ὅσαπερ σημαίνει τὰ σχήματα τῆς κατηγορίας, „to be in their own right are said
the things that are signified by the figures of predication“.
28 We may note that ἔξω is used here differently from how it is used in E.4 1028a2 (passage
[13]), this K.8 use spoiling the insight that accidental being and being as the true are parasital
upon categorial being.
Oddly, Jaeger ([see n.16], p.217) takes the rather casual way in which
the A7/E.2 account of the πολλαχῶς of being makes its appearance in K.8 as
corroborating the view that K.8 (that is, its E.2–4 part) was written earlier than
E.2–4: but what to say on this except that it, if anything, is a case of putting
things upside down?
V
Let us now pause for a moment and draw together what has come forth in
the two previous sections, adding a few appropriate remarks and expansions.
(α) To begin with, the Δ.7 approach to the πολλαχῶς of being, taken up
in E.2 init. (passage [10]), is quite different from the Γ.2 approach in terms of
a πρὸς-ἕν relationship (passages [5], [6]). In this case Aristotle did little apart
from simply transferring to being what he had observed regarding the things
that are called healthy. In contrast, the Δ.7 approach is based on the various uses,
or senses, of είναι, ‘to be’: it may mean
– to be F accidentally, κατὰ συμβεβηκός, as, e.g., for Socrates to be
musical, yielding accidental compounds such as musical Socrates, in
contradistinction to
– to be F essentially, καθ᾽ αὑτό, as, e.g., for Socrates to be a man, for
whiteness to be a colour, and so on; to be essentially is equivalent to
be an item in a category, and the categories in turn are determined by
‘the figures of predication’, σχήματα τῆς κατηγορίας, such that there are
as many ways of καθ᾽ αὑτὸ εἶναι as there are such σχήματα (see Δ.7
1017a22f., quoted in n.26).
– furthermore, there is the alethic use of εἶναι, according to which ἔστι
means ‘is true’, as we have said above (after passage [12], and see n.25),
yielding being as the true; and again,
– here is the potential construction of εἶναι, according to which it means
to be F potentially, or to be F actually.
In this much more complex configuratation not only categorial being is
sharply demarcated, but show up also ways of being that do not fit into the
πρὸς-ἕν structure of Γ.2 at all, viz., accidental being and being as the true. For
the πρὸς-ἕν structure of being fits best the case when one thing is predicated
of another, e.g. the quality musicalness of Socrates, the privation blindness of
Homer, the negation not-man of the God Zeus, and what not. But obviously
musical Socrates cannot be predicated of Socrates, nor the being as the true
Socrates is musical: these items relate quite differently to the entities they indicate,
in that these are already contained in categorial being such these two modes of
being are parasitical upon the latter.
The Δ.7 structure, however, does not exclude the πρὸς-ἕν structure but
includes it, in that it obtains within categorial being, as the Z.1 part of [9] shows.
Thus the structure of being according to the Δ.7 approach may be represented by
the following diagram:
accidental being – being as the true

parasitical upon

categorial being mapped onto it (cf., e.g., Θ.10 init.): potential and actual being
  προς-εν relationship to
substance.
And it goes without saying that this more complex Δ.7 scheme is the more
mature account of the πολλαχῶς of being than the simple πρὸς-ἕν structure of
Γ.2, providing as it does the frame for the entire discussion from E.2 to book Θ.
Thus, all those who from the fact that K.1–8 presents a parallel treatment
of the content of BΓE with K.3 and K.8 corresponding to Γ.1–2 and to E.2–4
respectively, infer that book E is to be tied not to the Substanzbücher but to ΒΓ,
simply ignore the deep-going difference between the Δ.7 scheme taken up in E.2
init. and the more simple account of the πολλαχῶς of being in Γ.2 init.; they put
the cart before the horse, for it ought to be asked whether in this essential point
there is a continuity from Γ to E in the first place: and since there is none, an
appeal to K.1–8 in order to attach E to ΒΓ is just idle.
(β) We have drawn attention to the fact that according to Δ.7 there is a
twofold πολλαχῶς of being, of which, however, the second, the πολλαχῶς of
categorial being (see passage [11]), is not mentioned in E.2. Again, this difference
is often ignored, and not taking it into account is the very basis of what seems
to be Jaeger’s main argument for the original disconnectedness of E and Z. Book
Z starts, to repeat, with saying that being, by now confined to categorial being
already, λέγεται πολλαχῶς, with a reference to Δ.7:
[17] Being is spoken of in many ways, as we distinguished them earlier in
the work on the various ways [in which things are spoken of] (τὸ ὂν λέγεται
πολλαχῶς, καθάπερ διειλόμεθα πρότερον ἐν τοῖς περὶ τοῦ ποσαχῶς).
For it signifies on the one hand what a thing is and this-something [i.e.,
substance], on the other hand of what quality or quantity [it is] or each of the
other things predicated in this way (σημαίνει γὰρ τὸ μὲν τί ἐστι καὶ τόδε τι, τὸ δὲ
ποιὸν ἢ ποσὸν ἢ τῶν ἄλλων ἕκαστον τῶν οὕτω κατηγορουμένων) (1028a10–13).
Now, Jaeger’s claim is that in case Z were written subsequently after E.2–4,
at the beginning of Z Aristotle would have referred back to E.2 but not to Δ,
which in that time certainly was a separate work περὶ τοῦ ποσαχῶς.29 But this
29 “Seltsamerweise fängt das neue Buch [Z] mit fast den gleichen Worten und derselben
Aufzählung der Bedeutungen des Seienden an, die unmittelbar vorhergeht: ‘Vom Seienden
spricht man in mehrfacher Bedeutung, wie wir früher – nun erwarten wir wenigstens einen
Rückweis auf die vorangehende Aufzählung in E 2, aber es folgt eine Überraschung – in
der Schrift über die mannigfaltigen Bedeutungen der Begriffe unterschieden haben. Denn
es bedeutet bald ein Etwas und ein bestimmtes Dieses, bald ein irgendwie beschaffenes oder
irgendwie groses oder ein beliebiges anderes aus der Anzahl dieser Art von Kategorumena’.
Hier ist ganz klar: wäre E 2 schon vorangegangen, als Aristoteles diesen Anfang des
Substanzbuches niederschrieb, so hätte er sich entweder auf die dort ausführlich entwickelten
verschiedenen Bedeutungen des Seienden berufen oder er hätte uberhaupt keine neue
Aufzahlung gegeben, da jeder sie ja im Gedächtnis hatte. Ist aber Z eine unabhängig von
argument of Jaeger’s is easily answered. In E.2, passage [10], the πολλαχῶς of

categorial being is not mentioned at all: indeed, this would have been far too early
there, where Aristotle sets out to discuss the first two members of the fourfold
πολλαχῶς, viz., accidental being and being as the true; only after they have been
dealt with, that is, towards the end of E.4, categorial being comes into focus, and
now it was fitting to recall that it too is a πολλαχῶς λεγόμενον, as is done in Z.1
init.: but now the reference can only go beyond E.2 to Δ.7 itself, where categorial
being is expressly said to be such, see passage [11]. And moreover, Aristotle
makes it perfectly clear what exactly he is referring to in Δ.7: after passage [17]
he adds in Z.1,
τοσαυταχῶς δὲ λεγομένου τοῦ ὄντος [...] a13f., „but while being is spoken
of in as many ways [...]”,
and this can only be taken as a deliberate allusion to Δ.7 1017a24,
τοσαυταχῶς τὸ εἶναι σημαίνει, „in as many ways [as there are figures of
predication] ‘to be’ signifies”.
So when Heidegger said that so far it has been overlooked, and in particular by
Jaeger, that the πολλαχῶς of being is twofold,30 he certainly had this argument of
Jaeger’s in mind. (Another case of this mistake we will meet in the final section VI).
(γ) An all-important juncture in the line of thought in book E is when at the
end of E.1 Aristotle identifies theology, the science of supreme being, with the
science of being qua being, i.e., being in general, which provides the frame for
the entire discussion in the rest of E, and in the Substanz-bucher ZHΘ.
So far we have, at the beginning of sec. IV, only in passing stated the mere
fact of this identification: but obviously more has to be said on it, since without
its soundness book E itself will fall apart into E.1 and the rest. Hence on this
point we will have to be fairly extensive.
To repeat, at the beginning of E.1 Aristotle takes at his starting-point the
conclusion of Γ.1, viz., that the enquiry is about the causes and principles of
being qua being (see [4], and the Γ.1 quotes in n.18). However, after explaining
(though more extensively than it had been done in Γ.1) the nature of this science
of being qua being, E.1 proceeds on lines quite different from those in Γ.
While in Γ, chapter 2 proceeds with stating that being is a πολλαχῶς
λεγόμενον displaying a πρὸς-ἕν structure, E.1 in contrast switches from the
den übrigen Metaphysikbüchern entstandene Untersuchung über die οὐσία, so ist gleich zu
verstehen, warum zu Beginn einer solchen die Stellung der οὐσία innerhalb der Gesamtheit
der moglichen Seinsaussagen zunächst an Hand der Kategorientafel kurz bestimmt werden
muste. Zu diesem Zweck berief Aristoteles sich auf die zweifellos öfter von ihm vorgetragenen
Ausfuhrungen Περὶ τῶν πολλαχῶς λεγομένων, [...] das sog. Buch Δ“. Jaeger [see n.16], p.
210. By and large, Frede & Patzig [see n.4], vol. II, p.10, seem to agree.
30 “Man hat diese Doppelung des πολλαχῶς bisher ubersehen; so besonders W. Jaeger“. Martin
Heidegger, Die Grundbegriffe der antiken Philosophie, Frankfurt am Main: Klostermann, 1993,
p.291. Of course, in this generality Heidegger’s statement is not true: after all, that Doppelung
is clearly expounded in Brentano’s dissertation of 1862, which, by the way, Heidegger
was very well acquainted with: see Franz Brentano, Von der mannigfachen Bedeutung des
Seienden nach Aristoteles, reprint Hildesheim: Olms, 1960, p.6f., 72. Thus it is very likely that
Heidegger himself derived his appreciation of that Doppelung from Brentano.
delimitation of the science of being qua being to a classification of the theoretical

sciences, or philosophies (φιλοσοφίαι θεωρητικαί 1026a18f.), according to their
subject-matter. These are three:
– physics or natural science, ἡ φυσική [sc. ἐπιστήμη], subject-matter: the
things which are ‘separable’,
i.e. existing independently, but not immovable (χωριστὰ μὲν ἀλλ᾽ οὐκ
ἀκίνητα 1026a14), that is, the sensible substances;
– mathematics, ἡ μαθηματική, subject-matter: things which are immovable
but perhaps not separable (ἀκίνητα μὲν οὐ χωριστὰ δὲ ἴσως a15); and
– divine science or theology, ἡ θεολογική, subject-matter: things which
are both separable and immovable (καὶ χωριστὰ καὶ ἀκίνητα a16), this
science being, in virtue of the dignity of its subjectmatter, the first one,
ἡ πρώτη ἐπιστήμη (a15f., a29), and that is, it is the First Philosophy, ἡ
πρώτη φιλοσοφία (a24).
And then, in the remaining part of E.1, Aristotle returns to the science of
being qua being and poses the question whether First Philosophy, i.e. theology,
having a particular subject-matter, is also that most universal science dealing
with being in general.
His answer is in the affirmative: if there were only natural substances as the
only things that exist independently, physics would be First Philosophy; but if
there is an immovable substance, the science dealing with it, i.e. theology, will
be First Philosophy, and just because of this the universal science of being qua
being as well:
[18] One might indeed be puzzled by the question whether First Philosophy
is universal, or deals with some one genus and some single nature of things
(ἀπορήσειε γὰρ ἄν τις πότερόν ποθ᾽ ἡ πρώτη φιλοσοφία καθόλου ἐστὶν ἢ περί τι
γένος καὶ φύσιν τινὰ μίαν) [...]
[We answer:] now if there is not a substance other than those formed by
nature, physics would be the first science (εἰ μὲν οὖν μὴ ἔστι τις ἑτέρα οὐσία παρὰ
τὰς φύσει συνεστηκυίας, ἡ φυσικὴ ἂν εἴη πρώτη ἐπιστήμη): but in case there is an
immovable substance, this one is prior and [the science of it] is First Philosophy,
and general in this way, just because it is first (εἰ δ᾽ ἔστι τις οὐσία ἀκίνητος, αὕτη
προτέρα καὶ φιλοσοφία πρώτη, καὶ καθόλου οὕτως ὅτι πρώτη) – and it will
belong to this science to study being qua being, both what-it-is and what holds of
it qua being (καὶ περὶ τοῦ ὄντος ᾗ ὂν ταύτης ἂν εἴη θεωρῆσαι, καὶ τί ἐστι καὶ τὰ
ὑπάρχοντα ᾗ ὄν) (Ε.1 1026a23–32).
Aristotle’s answer to the question posed, namely, that the science of the
highest kind of being is by this very fact also the science of being in general, has
perplexed commentators. Bonitz [see n.6], p.285: licet in ipsis verbis difficile sit
solutionem agnoscere.
Jaeger even saw in this answer a blatant contradiction. He held the view that
in Aristotle there are two entirely different, and incompatible, conceptions of
metaphysics or First Philosophy, viz., an earlier, more Platonic one, according to
which it is theology and to which ABΓ + Ε.1 without the concluding passage [18]
belong, and a later, genuinely Aristotelian one, which Aristotle expounds in the
Substanzbücher and according to which metaphysics is the general investigation
of substance, or more precisely, of sensible substance: only later, when he joined
the originally independent Substanzbücher to the earlier treatise ABГE.1, Jaeger
argues, Aristotle wrote Ε.2–4 as connecting piece and added [18] to Ε.1 as a
marginal note, which, however, makes in his view the clash between the two
irreconcilable conceptions of metaphysics even more glaringly obvious.31
Such criticism is, however, entirely misconceived. The science of being qua
being is to study the principles and causes (or rather, the ‘becauses’, τὰ διότι)
thereof, and thus also, as Aristotle says in the last line of Ε.1, the τί ἐστι, the
What-it-is or quiddity, and that is, the essence or the What-it-is-to-be-something,
the τί ἦν εἶναι, of things, i.e., their formal cause. Now there are no causes, and
hence no essence, to be studied of what does not exist, see, for instance, An. Post.
II.8 93a18–20:
[19] It is not possible to grasp the Why before the That, and in the same
way it is plain that the What-it-is-to-be-something is not without the That (οὔτι
πρότερόν γε τὸ διότι δυνατὸν γνωρίσαι τοῦ ὅτι, δῆλον ὅτι ὁμοίως καὶ τὸ τί ἦν
εἶναι οὐκ ἄνευ τοῦ ὅτι ἐστίν): for it is impossible to know what a thing is, if we are
ignorant whether it is (ἀδύνατον γὰρ εἰδέναι τί ἐστιν, ἀγνοοῦντας εἰ ἔστιν) (cf.,
e.g., II.7 92b4–8).
Or in Aquinas’ measured words: quaestio enim quid est sequitur quaestionem
an est.32 And because of this existential presupposition it is plain that in case
there are only sensible i.e. movable substances, physics will be both First
Philosophy and the science of being qua being: but if there is an immovable
substance, theology as the study of that substance which then turns out to be
the first cause of motion in the universe, τὸ πρῶτον κινοῦν ἀκίνητον αὐτό (Γ.8
1012b31, Л.7 1072b7), will also encompass per primam causam, so to speak, the
domain of the movable, and will thus be the science of being qua being as well.
Thus the only presupposition for the identification of theology with the science
of being qua being is that the universe does not fall apart into mere episodes
like a bad tragedy (as Aristotle puts it in N.3 1090b19f.), i.e., that the domains of
theology and of physics are interconnected, as they are by the relationship of first
mover to what is moved by it.33
Thus it would seem that our problem with the concluding passage of E.1 is
rather our being prejudiced by modern philosophy: the study of being qua being,
31 “Die Randglosse schafft den Widerspruch nicht fort, sie macht ihn im Gegenteil nur noch
sichtbarer [...] Die beiden Ableitungen des Begriffs der Metaphysik sind zweifellos nicht
aus ein und demselben geistigen Schöpfungsakt hervorgegangen. Zwei grundverschiedene
Gedankengänge sind hier ineinandergeschoben. Man sieht sogleich, das der theologisch-
platonische der ursprünglichere und ältere ist, [...] während umgekehrt die Definition des
ὂν ᾗ ὄν [...] der letzten und eigenartigsten Entwicklungsstufe des aristotelischen Denkens
[entspricht]”. Jaeger [see n.16], p.227.
32 S. Thomae Aquinatis in octo libros Physicorum Aristotelis Expositio. Ed. P. M. Maggiolo, Turin;
Rome: Marietti, 1965, lib. IV, lect. 10, §507.
33 It seems that Kirwan views the matter in a similar way: see Aristotle. Metaphysics Books Γ, Δ,
and E. Translated with notes by Christopher Kirwan, Oxford: Clarendon Press, 21993, p.188f.
then called ontology, is in the first place study of essence, and since Descartes
turned Aquinas’ dictum upside down by saying that
iuxta leges verae Logicae de nulla unquam re quaeri debet an sit, nisi prius
quid sit intelligatur, 34
it is simply taken for granted that essence is daseinsfrei, or at least can be
studied without bothering whether it is instantiated.35 But as soon as we take
seriously the fact that Aristotelian essence is in the way stated entirely different
from modern conceptions of essence, it will perhaps be not so difficult anymore
to appreciate Aristotle’s move in the concluding E.1 passage.
Of course, one might also raise the more basic question why Aristotle
did not move from the beginning of E.1, where being qua being was the issue,
straightforwardly to the πολλαχῶς of being which we have at the beginning
of E.2 in the first place, but went on with classifiying the theoretical sciences
to finally identifying the supreme one of them with the science of being qua
being and thus returning only after a long detour in a sort of Ringkomposition,
as classical scholars call such a thing, to the starting-point. A possible answer
could be this. Aristotle’s intention in the treatise de essentia et de ente was to
eventually study non-sensible substance, as he frequently says in it: 36 but since it
has now become clear that the supreme science of non-sensible substance is in a
certain way a science of sensible substance as well, he is justified in studying first
sensible substance, what he does in ZHΘ, as a sort of preparation for the study
of non-sensible substance which would have brought the treatise de essentia et de
ente to completion.
VI
We have, in some detail, considered arguments that have been brought forth
for severing book E from the Substanzbücher and attaching it to the books ABГE
and found them wanting in cogency; and simultaneously good evidence has
shown up for the view defended here, namely, that in E the discussion of the
science of being qua being is taken up not only quasi de integro, as Bonitz had
put it, but de integro simpliciter, i.e., Aristotle makes a fresh start in his efforts at a
systematic exposition on that subject, which he completes up to Θ and which he
then in I.2 refers to as the treatise περὶ τῆς οὐσίας καὶ περὶ τοῦ ὄντος, de essentia
et de ente.
We have followed the consistent thread of argument in book E from its very
beginning, saying that
34 Meditationes: Primae Responsiones. Descartes, OEuvres. Ed. Ch. Adam & P. Tannery, vol. VII,
Paris: Vrin, 1964, p.107f.
35 Thus our Graz celebrity Meinong found it even worthwile to formulate, in §3 of his 1904
essay Uber Gegenstandstheorie, a special principle of independence of essence from existence,
which he called „das Prinzip der Unabhangigkeit des Soseins vom Sein“. Alexius Meinong,
Uber Gegenstandstheorie. Selbstdarstellung, Hamburg: Meiner, 1978, p.8.
36 See, e.g., Z.2 1028b28–32, Z.3 1029b3–12, and Z.17 1041a7–9.
– the object of inquiry are the principles and causes of being qua being
(passage [4]),
through
– the identification of the science of supreme being with the science of
being qua being in the concluding passage [18] of E.1,
– the return to the topic of being qua being in terms of the fourfold
partition of being according to the Δ.7 scheme at the beginning of E.2
(passage [10]),
the result of the subsequent discussion first of accidental being, and then of
being as the true, viz.,
– that both accidental being and being as the true are to be excluded from
the domain of the science of being qua being,
stated towards the end of E.4, up to the conclusion that
– the object of inquiry, σκεπτέον 1028a3, are the principles and causes of
categorial being. And this is where Z.1 takes up the discussion, passage [17].
There remains to consider the transition from E to Z itself. After the
σκεπτέον conclusion 1028a3, E has one more sentence:
[20] But it is clear in the work in which we distinguished the various ways
in which things are spoken of, that being is spoken of in many ways (φανερὸν δ᾽
ἐν οἷς διωρισάμεθα περὶ τοῦ ποσαχῶς λέγεται ἕκαστον, ὅτι πολλαχῶς λέγεται τὸ
ὄν) (E.4 1028a4–6).
This last sentence of book E says in quite similar words the very same thing
as the next one which book Z begins with, to put it from passage [17] down once
more:
Being is spoken of in many ways, as we have distinguished them earlier in
our work on the various ways [in which things are spoken of] (τὸ ὂν λέγεται
πολλαχῶς, καθάπερ διειλόμεθα πρότερον ἐν τοῖς περὶ τοῦ ποσαχῶς) (1028a10f.).
Regarding the last sentence of E Ross ([see n.1], vol. I, p.366) says:
The remark is pointless here, as it has already been noted (1026a33 [E.2 in.])
that ‘being’ has a variety of meanings and two of them have been discussed in chs.
2–4. The sentence is a free version of the first sentence of Z, and is evidently a
later addition meant to indicate the connexion of the two books.
But of course, the last sentence of E is not pointless at all: just as Jaeger, Ross
does not appreciate that before in E.2 only the fourfold πολλαχῶς of being was
mentioned without any indication that the third member of that partition, viz.,
categorial being which is focussed on by now, itself is a πολλαχῶς λεγόμενον too.
Now, one of the two sentences is superfluous. On Bonitz’ view ([see n.6],
p.14, 294), it is a matter of indifference which one we retain. Following Schwegler’s
proposal ([see n.7], vol. IV, p.33), we delete the second one from the beginning of
Z. Then the whole passage from E.4 1028a3 to Z.1 1028a13 reads as follows:
[21] These two [accidental being and being as the true] are to be dismissed
(ἀφείσθω), but what what is to be investigated (σκεπτέον 1028a3), are the causes
and principles of being itself qua being. But it is clear (φανερὸν δ᾽ a4) in the work
in which we distinguished the various ways in which things are spoken of, that
being is spoken of in many ways.
For it signifies (σημαίνει γάρ a11) on the one hand what a thing is and this-
something, on the other hand of what quality or quantity [it is] or each of the
other things predicated in this way.
No superfluous repetition and no disrupting separation of books conceals
anymore the uninterrupted flow of thought from E to Z: it is all of one piece.
And to add, the problem with the connective particle missing at the beginning
of Z.1 has vanished, just as another awkwardness has gone: the word πρότερον,
‘earlier’, in the first sentence of Z.1 which, as Frede & Patzig ([see n.4], vol. II, p.10)
note, Aristotle normally uses to refer to an earlier discussion in the same work.
Afterwards in Z.1, Aristotle argues that substance is among categoral being
primary, which here we need not go into any further anymore, and concludes
Z.1 with the words that
[22] we must investigate chiefly and primarily and so to speak exclusively
what that is which is being in that sense [i.e. substance] (ἡμῖν καὶ μάλιστα καὶ
πρῶτον καὶ μόνον ὡς εἰπεῖν περὶ τοῦ οὕτως ὄντος θεωρητέον τί ἐστιν) (1028b6f.).
Pay attention to the θεωρητέον 1028b7, ‘it is to be investigated’: it corresponds
exactly to the σκεπτέον 1028a3 towards the end of E.4. That σκεπτέον refers to
categorial being in its entirety as that what is to be studied by the science in
hand. This result was, however, only provisional. But now the θεωρητέον in the
last line of Z.1 has finally pinned down the most proper object of this science,
namely substance.
In this way, then, Z.1 completes in one continuous exposition what book E
is all about, and to this extent we can fully agree with Schwegler’s objection to
the separation of E from Z. But only this far, for then in Z.2 Aristotle does what
he usually is doing when starting an enquiry into some topic, namely, to survey
first various opinions about it embedded both in common sense and in relevant
scientific theories, which are in this case, among others, the doctrines of Plato
and of Speusippus in particular. (These two are mentioned Z.2 1028b19–21): and
so one may conclude that Z.1 still belongs to book E, and that book Z really
begins with Z.2.
In closing we may look back at / .1–8, and suggest the following alternative
hypothesis about it. When Aristotle prepared the course on First Philosophy
of which K.1–8 may well be a pupil’s notes, he used for it texts he already had
written some time ago, i.e. Β and Γ, but as well what he at that time was just
working on, viz., the treatise de essentia et de ente which as work in progress had
got at least as far as to what we have as Z.1: and this required him to adapt in that
most crucial point of the πολλαχῶς of being the Γ.2 list of being to Z.1, such that
the K.3 list resulted which bears that striking resemblance to Z.1 1028a18–20
which we observed when in section III (see [9]) we compared the two passages,
and which cannot possibly be explained by the Γ.2 approach to the matter.
Sanja Srećković Original Scientific Paper
University of Belgrade UDK 111.852
Ханслик Е. 78.01
EDUARD HANSLICK’S FORMALISM AND HIS

MOST INFLUENTIAL CONTEMPORARY CRITICS
Abstract: The paper deals with the formalistic view on music presented in Eduard
Hanslick’s treatise On the Musically Beautiful, which is taken to be the founding
work of the aesthtetics of music. In the paper I propose an interpretation of
Hanslick’s treatise which differs on many points from the interpretations displayed
in the works of several most influential contemporary aestheticians of music. My
main thesis is that Hanslick’s treatise is misunderstood and incorrectly presented
by these authors. I try to demonstrate this thesis by referring to Hanslick’s original
formulations in the German edition and by showing that my interpretation renders
Hanslick’s view far more coherent and his arguments successful in showing his
main conclusions. Accepting this alternative interpretation should have further
implications on many contemporary theories in the aesthetics of music that reckon
on the failure of Hanslick’s arguments as presented by usual interpretations.
Key words: aesthetics of music, Eduard Hanslick, formalism
Introduction
Eduard Hanslick is considered the founding father of the aesthetics of
music. His treatise On the Musically Beautiful1 has had such a great impact on
the understanding of music, that today, almost all of the authors who write about
music from a philosophical perspective, especially about music’s relation to
emotions, first state their attitude toward Hanslick’s view, and only then present
and defend their own view. In contemporary literature on music, a distinction
is made between the so-called Hanslickians and Hanslick’s critics, although all
of them mostly reject all of his arguments. In this paper, I shall not deal with
contemporary authors’ attempts to establish their own theories. Rather, I will
focus on their interpretations of the treatise On the Musically Beautiful, since,
in my opinion, that is where a more fundamental, and also more interesting
point can be found. Namely, I will argue that Hanslick’s treatise (or „little book”
1 I will comparatively follow Geoffrey Payzant’s translation: Eduard Hanslick, On the Musically
Beautiful, trans. Geoffrey Payzant, Indianapolis, 1986. (in further text – OMB), for it is this
translation that contemporary authors refer to, and Ivan Foht’s translation: Eduard Hanslik, O
muzički lijepom, pr. Ivan Foht, BIGZ, 1977. (in further text – OML), because it is more faithful
to the original than Payzant’s, which partly engages in interpreting the text. In addition, I
will provide parts of the original text when mentioning contentious or key terms: Eduard
Hanslick, Vom Musikalisch-Schönen, 13.–15. Auflage, Leipzig, 1922. (in further text – VMS).
as it is called by some of these authors) is superficially read by these authors,

and mostly misunderstood. This fact should reflect considerations that follow
from the mentioned misunderstanding. In this paper, I shall offer a more
precise reconstruction of Hanslick’s argumentation. After that, I shall present
interpretations of Hanslick’s view by several most influential contemporary
authors, followed by a demonstration that there is a discrepancy between these
interpretations and the text of the treatise. The aim of the paper is to prove a
possibility of a different interpretation of Hanslick’s treatise, which I consider not
only to correspond to the text of the treatise better, but also to present it as more
coherent than other interpretations do.
Firstly, I will try to explain the so-called negative thesis, which is what
Hanslick is most occupied with in his treatise. Explicitly, Hanslick only writes
that his negative thesis „first and foremost opposes the widespread view that
music is supposed to ‘represent (darstellen) feelings.’„2 In elaboration and
demonstration of this thesis, Hanslick will examine other relations of music and
feeling, relations that will be more interesting to the contemporary aestheticians
of music. Contemporary authors do not follow Hanslick’s line of argumentation,
but rather directly go in for the assertions and conclusions that they are interested
in. However, I will here hang on to Hanslick’s course of exposure, since I believe
his work to be far more coherent than it seems at first, and also that neglecting
the relations between certain arguments easily leads to misunderstanding of
Hanslick’s assertions and intentions.
While the explicit statement of the negative thesis does not say much, broadly,
the thesis includes critique and rejection of then widespread apprehension of
music, which Hanslick ironically names „the emotional aesthetics.” It is a view,
which, according to Hanslick, attributes a double role to emotions in the art of
music: firstly, that the arousal of emotions is what defines music and the purpose
of music (Zweck und Bestimmung); secondly, that emotions are the content of
music, „that which musical art represents in its works.”3 Hanslick rejects both of
these roles, and his arguments against them constitute the negative part of the
treatise. In the next section, I will discuss what exactly Hanslick means by these
two roles and how he dismisses them. First, I shall consider how the definition
and purpose of music should be understood.
The first role: (aroused) as the defining purpose of music

Considering this role, music is understood from a functionalist point of
view: fulfilling a certain unartistic function or purpose (such as the arousal of
emotions) is the specificity of music – it is what music does and what determines
it as music. At the same time, fulfilling of this purpose is what music is supposed
2 See OMB, p. xxii, OML, p 36–7. I will accept represent the correct translation of the term
darstellen. Hanslick explicates this term as following: to represent is to produce a clear and
distinct content, to «put» (or present) it «before» our very eyes (»daher stellen«) – see OMB,
p. 14, OML, p. 66. Contrary to Payzant’s translation of this term as represent, contemporary
authors mostly speak of expression or expressiveness, which Hanslick names as ausdrücken
throughout the whole treatise.
3 See OMB, p. 3, OML, p. 41–2.
Sanja Srećković: Eduard Hanslick’s formalism and his most infuential contemporary critics 115
to do, a task it has to accomplish. This, as it turns out, descriptive-normative

definition (since it includes what music de facto does, as well as what it should
do) is not formulated explicitly within some elaborate theory. It was only a part
of a then widespread way of thinking about music, and about art in general.
Hanslick does not go into details of such thinking (which he considers to be very
vague), but rather dismisses it completely.
Hanslick first rejects the normative component of the definition: he claims
that beauty has no purpose at all, for beauty is mere form, i.e. a shape in which
something appears, which by itself has no purpose.4 Although something that
is beautiful can be applied to many purposes, and even used solely for a specific
purpose, this purpose is not what makes the thing beautiful, i.e. the thing is not
beautiful because some intention or purpose is realized.5 Beauty, in Hanslick’s
view, is in the thing that is beautiful, and which remains beautiful „even if no
feelings are aroused and even if it be neither perceived nor thought.”6
Stating that the arousal of emotions is what defines music means that arousing
human emotions is specific for music, and differentiates it from other arts.
Hanslick’s objection is simple: emotions are aroused not only by other arts, but also
by various non-aesthetical preoccupations (Hanslick’s examples are rhetoric and
religious fervour), and by everyday events as well.7 Hanslick does not see how can
in principle music be differentiated from other arts by arousal of emotions.8
Furthermore, although works of arts do stand in some kind of relation to
our emotions, none of them stands with in an exclusive relation to emotions. In
other words, art is not exhausted in its connection with human emotions.9 Is it
only one of many aspects of music, and therefore, insufficient for defining it.
4 It is easier to understand this statement in connection with the explanations Hanslick gives
later in the text, when he compares artistic and natural beauty or when he compares music
with arabesque and kaleidoscope: the beauty in all those examples is in the form itself,
independent of what is formed or shaped. This means that beautiful objects are beautiful due
to their formal properties, that is, the properties accessible by sensory perception. In the case
of both artistic and non-artistic objects, beauty concerns the order or harmony of the parts of
the object, while any possible purpose of the object does not affect its beauty.
5 See OMB, p. 3, OML, p. 42. Hanslick will explain what does make a beautiful thing beautiful
in the positive thesis.
6 Ibid.
7 See OMB, p. 5, OML, p. 45.
8 One possible answer that Hanslick mentions is that the difference is in the manner of arousal:
that music arouses emotions directly, as opposed to other arts where the arousal is mediated
by conceptual content. However, Hanslick objects to this by explaining that the arousal
caused by listening to music is also multiply mediated starting with sensations, imagination
and so on. Feeling is nothing more than a secondary effect of music – as well as of other arts
– see OMB, p. 4, 5–6, OML, p. 43, 45–46. Music does stand out in its effect on the feelings
because it affects us more rapidly and intensely than other arts. Hanslick demonstrates
further in the treatise that this difference is physiologically conditioned and does not arise
from the artistic aspect of music, but rather from its material that is in a certain relation with
our physiological organisation. Therefore, this effect cannot be something that is artistically
specific of music – see OMB, p. 50–1, OML, p. 118–9.
9 See OMB, p. 6, OML, p. 46.
Hanslick proposes additional reasons against such defining, which

demonstrate how uncertain and variable the relation between music and the
emotions aroused by it is. Firstly, our thoughts and emotions elicited by music
are often guided by titles, texts and other merely incidental associations of ideas
(especially in church, military and theatre music), which we often wrongly
ascribe to the music itself. Besides, the relation between an art of music and
our mood changes is also determined by our changing musical experiences and
impressions. Hanslick asserts that nowadays we can scarcely understand how
some earlier generations of music listeners could regard some particular music
sequence as a precisely corresponding expression (Ausdruck) of a particular
feeling. He supports this assertion by citing „the extraordinary difference
between the reactions of Mozart’s, Beethoven’s and Weber’s contemporaries
to their compositions in different historical periods. He concludes that the
way feelings will be aroused by music depends on the circumstances of each
particular instance.10
After this, Hanslick makes an additional remark, that throughout this
variation in the reactions to music, the musical value of the works themselves
remains unaltered. This leads Hanslick to conclude that the effect of music upon
feelings possesses neither the necessity nor the constancy nor the exclusiveness
(in relation to the value of the musical work) which a phenomenon would have
to exhibit in order for us to base the criteria for work’s value on it. Thus, Hanslick
rejects not only the view that the aroused emotions define music, but also the
view that the arousal of emotions can be the criteria for value (i.e. aesthetical
principles) of musical works.
The second role: emotions as the content

(which should be) represented by music
From the differences among human senses, to which the arts are bound,
follow the fundamental differences in the way the various arts shape their
products (i.e. the difference of their forms), and from that, further, follows the
diversity of content among the arts. Every particular art has its own range of
ideas which it represents in its own form of expression, with its own means of
expression (Ausdrucksmitteln), e.g., tones, words, colours, etc. Hanslick claims
that an artwork embodies a certain idea as beauty in sensuous appearance.11 In
other words, by these means of expression an artist externalizes or embodies
an artistic idea in such a way that the idea appears to our senses as something
beautiful, as a beautiful form.
The view that Hanslick rejects is the belief that the whole range of human
feelings is the content of music, that these feelings are the ideas music acoustically
10 See OMB, p. 6–7, OML, p. 47–8. Hanslick also notes that, while Mozart’s music was described
by his contemporaries as passionate and struggling as opposed to the tranquillity and
wholesomeness of Haydn’s works, the same comparison was made just a few decades later
between Mozart and Beethoven.
11 See OMB, p. 8, OML, p. 55.
embodies in shaping its works. According to this view, tones and their artistically
shaped relations would be mere raw material, the means of expression by which
the artist represents human feelings. Hence, what we enjoy in music and what
has an effect on us is not the melody and the harmony we are listening to, but
the feelings they signify (bedeuten).12
Hanslick’s statement, which will be under many attacks in later philosophical
works, is the claim that music is not able to represent definite feelings.13 In order
to demonstrate that, he proposes the following arguments:
In the first argument, Hanslick claims that what is specific for a feeling, what
renders it a definite feeling, and hence is the criterion by which we are able to
extricate and discern it from other feelings, is its conceptual content, a number
of conceptions and judgments (Vorstellungen und Urteile) which are associated
to the concept of that feeling. Therefore, e.g., „the feeling of hope cannot be
separated from the representation of a future happy state which we compare with
the present,” or „love cannot be thought without the representation of a beloved
person, without desire and striving after felicity,” etc.14 Without this conceptual,
cognitive apparatus, all that remains is an unspecific stirring, a general state of
well-being or distress, and the intensity and oscillations which are not specific
for the feeling (because different feelings can have the same intensity and
oscillations, and, also, the same feeling can vary in its intensity). It is not clear
how pure instrumental music could be able to convey or reproduce concepts
or representations, for they are not within the scope of music.15 Since the
specification of the feelings cannot be separated from their conceptual content,
and since music cannot reproduce (wiedergeben) concepts, Hanslick concludes
that music cannot represent specific feelings.16
The second argument states that the only aspect of feelings music is capable
of representing, since it is the aspect they both share, is the dynamic aspect, the
dynamic flow of experiencing a feeling. Music can imitate (nachbilden) only
the motion of a physical process, the characteristic way of motion. Motion is
something both music and feelings share, but it is only one aspect of feeling,
which, as said already, can be shared by different feelings, and also vary in one
feeling, since the same feeling does not always have the same dynamics. Hence,
dynamics cannot be specific for a feeling.17 Therefore, since the only aspect of
feelings which music can represent is not specific for any feeling, it follows that
music cannot represent specific feelings.
The argument I propose as the third one emphasizes that the ideas of
specific feelings cannot appear or be embodied in instrumental music because
12 See OMB, p. 8–9, OML, p. 56.

13 See OMB, p. 9, OML, p. 56, VMS, p. 22: „Die Darstellung eines bestimmten Gefühls oder
Affektes liegt gar nicht in dem eigenen Vermögen der Tonkunst.“
14 See OMB, p. 9, OML, p. 57.
15 See OMB, p. 10, OML, p. 58.
16 See OMB, p. 9, OML, p. 57–8.
17 See OMB, p. 11, OML, p. 60–1.
there is no necessary (notwendiger) connection between those ideas and certain

combinations of musical tones. I take this to mean that there is nothing in
the tones themselves and in their combinations which would necessarily or
unambiguously refer to feelings.18 This was just a statement Hanslick mentioned
within the argument concerning the dynamics. However, I identify it as a
separate argument because I take it as implicitly stating that, because of lacking
the so-called „necessary” connections, music cannot present its tone structures
as structures relating to feelings. It cannot present its dynamic flow as the flow
of a feeling.19 Therefore, even though music can successfully imitate certain
aspects of feelings, it has no means to connect those aspects unambiguously
with a feeling. In other words, the connection between music and feelings always
stays in our open interpretations.
In the forth argument, Hanslick appeals to the empirical testing of the thesis
that music can represent feelings. However, I consider that, at the same time, he
points out a certain weakness of the emotional aesthetics and all similar theories,
and that is the justification for ascribing any particular emotional content to an
artwork. Namely, says Hanslick, if someone considers that music indeed can
represent feelings, let him try to demonstrate, with arguments, which feeling
constitutes the content of an artwork.20 If we should ask the listeners of the most
popular musical works at the time, from Mozart to Chopin, to name the feeling
they consider to be the content of the work, there would, in Hanslick’s opinion,
be different answers (e.g. love, yearning, piety).21 As well as there could be no
demonstration of why any of the answers would be correct, nobody could refute
any of them, either.22 Therefore, there is not only the discrepancy about which
feeling is represented by some musical work, the fact is that we do not even have
at our disposal the means – the reasons, by which we could choose the right
answer to what is the content of the work.23
In the fifth argument, Hanslick turns to the representation of feelings in
vocal music as well. First of all, it should be noted that for Hanslick, the relation
18 Ibid.
19 Malcolm Budd later explicated this argument (See Malcolm Budd, Music and the
Emotions: The Philosophical Theories, London, 1985. p. 24.), and Steven Davies took this
explication over as an addition to the rest of Hanslick’s arguments which refutes all attempts
at crossing the gap between music and the emotions (see Stephen Davies, Musical Meaning
and Expression, Ithaca NY, 1994. p. 217–9.).
20 See OMB, p. 12, OML, p. 62–3.
21 Earlier in the treatise, Hanslick mentions different documented descriptions of the same
compositions and notes that the listeners have had different opinions about which sequence
of tones were adequate expressions for which feeling. See OMB, p. 6–7, OML, p. 47–8.
22 See OMB, p. 14, OML, p. 65.
23 See OMB, p. 14, OML, p. 65–6. The argument supposes Hanslick’s sense of representation as
exposition of a content in a manifest manner. Thus it is understandable that Hanslick does
not see how we can designate something as what an art represents, when the it is the very
dubious and ambiguous element of that art, perpetually subject to debate. Also, this lack of
reasons (which would support the ascription of any emotional content to a composition)
is connected to the previous argument, that is, with the lack of the necessary connection
between tone groups and definite feelings.
between music and the text in a vocal composition is such that music alone
with its assets has the power only to enhance, revive, or colour the content of
the text. But what is represented in a vocal work is determined by the text, not
by the music.24 Hanslick claims that music is not determined enough so that
it could represent feelings in a determinant way, thus the same melody might
just as appropriately be played along some words with a very different, even
the opposite meaning. Hanslick substantiates this with a number of examples.
Namely, concerning Gluck’s opera Orpheus and Eurydice, a contemporary of
Gluck remarked that, in the Orpheus aria (which moved many of the listeners to
tears), „one could just as well or, indeed, much more faithfully set the opposite
words to the same tune” (that instead of „I lost my Eurydice,” the words say „I
found my Eurydice”).25 Hanslick himself remarks in addition to this that „music
certainly possesses far more specific tones for the expression of passionate
grief ”26 and emphasizes that he picked this example intentionally, first, because
Gluck „has been atributed the greatest exactitude in dramatic expression and,
second, because over the years so many people have admired in this melody the
feeling of intense grief which it expresses in conjunction with those words.”27
What was true for these smaller parts of musical works, Hanslick continues,
applies to the whole compositions as well. Many whole vocal compositions have
used different texts with the same music. Hanslick lists a number of examples:
ouverture to Mozart’s The Magic Flute is often performed as a vocal quartet
of quarrelsome Jewish shopkeepers, and Mozart’s music, with not a single
note altered, suits the low-comedy words surprisingly well; in many German
provincial churches sentimental love songs are performed during the celebration
of the Mass; in Italian churches, also, popular wordly tunes of Rossini, Bellini,
Donizetti and Verdi are played, and they do not at all disturb the devotions of
parishioners, but, on the contrary, inspire them even more; many of the most
famous pieces in Hendel’s Messiah, including the ones most admired for their
godly sentiments, are taken from the secular and mainly erotic duets which
Handel composed earlier; the madrigal-like pieces in J. S. Bach’s Christmas
Oratorio were transcribed from altogether dissimilar secular cantatas; Gluck,
who is considered to has achieved great dramatic truth in his music only by
fitting each note precisely to the specific situation, in Armida transcribed five
pieces from his earlier Italian operas.28 These examples lead Hanslick to conclude
that if music itself were capable of representing a specific feeling, say, devotion,
24 See OMB, p. 16, OML, p. 67.

25 See OMB, p. 17, OML, p. 67–8.
26 See OMB, p. 18, OML, p. 69. I believe that this passage shows Hanslick’s attitude toward the
issue of music’s expressiveness. It is clear in the passage that he thinks that melodies and
other musical elements or passages can be suitable for the expression of feelings, but not in
a way in which they would be tied to the expression of only one particular feeling. The same
melody could express several different, even disparate feelings. In other words, Hanslick
takes musical elements to be indefinitely expressive.
27 Ibid.
28 See OMB, p. 18–19, OML, p. 69–72.
as its content, such a quid pro quo would be as impossible as if the preacher
could substitute his sermon with reciting a novel or a page of parliamentary
transactions.29
The beauty and the representation of feelings in music

After the exposition of the arguments which demonstrate that music cannot
represent specific feelings, Hanslick further considers the possibility that the
representation of feelings is an ideal for music which it can never achieve, but
which it can and should strive to approach.30 This would mean that the beauty
of music resides in the representation of feelings in the degree in which such
representation is possible. Hanslick proposes arguments which refute such a
possibility by taking the counterfactual point of view: let us imagine that musical
representation of definit feelings is possible so we could show that the degree of
exactitude with which music represents them does not coincide with the beauty
of the artwork.31 Hanslick tries this fiction with vocal and not instrumental
music, since instrumental music by itself automatically excludes the possibility
of identifying definite feelings, whereas in vocal music the feelings allegedly
represented are indicated by the text. The words in vocal music determine what
is being represented and what is being revived by music in tonal motion.32 If
we suppose that musical representation of feelings is possible and that the
exactitude of the representation coincides with the beauty of the artwork,
then we should consider those compositions which accomplish that task most
specifically (bestimmtesten) the most beautiful.33 However, firstly, everyone
knows of musical pieces we evaluate as works of utmost beauty which have no
such content (e.g. the preludes and fugues of J. S. Bach). Secondly, there are,
opposed to these, vocal compositions which try to portray a specific feeling
(within the limits just explained) and in which truth of portrayal has precedence
over any other principle, and, yet, which do not possess a great degree of beauty.
The third argument against the same thesis states that in a vocal composition
we can make many small alterations, which, without weakening the expression
29 See OMB, p. 18–19, OML, p. 70.

30 See OMB, p. 21, OML, p. 74.
31 Ibid.
32 See OMB, p. 21–2, OML, p. 75. Contemporary interpreters consider this transition to the
examples of vocal music as a problem for Hanslick, so I will try to further explicate this
passage and reconstruct what might have been Hanslick’s motivation for making that choice.
Namely, so far Hanslick was speaking solely about the instrumental music (since, as he
says, we may only say that music can do of what instrumental music, with no help from
non-musical means, can do), and he was demonstrating that it cannot represent definite
feelings. Now, for the sake of the argument, he supposes that the music in vocal pieces can
represent feelings (that is, he imagines that representing the dynamic aspect is sufficient for
the representation of definite feelings) which are the content of the text. The text enables us
to identify what is being represented by the music and we can thus judge to what extent was
the representation successful.
33 See OMB, p. 22, OML, p. 75.
of feeling, yet immediately destroy the beauty of musical themes. This would be
impossible if the beauty consisted in the exactitude of expression of feelings.34
Therefore, we can make changes in one aspect without affecting the other.
The supposed thesis cannot explain this mismatch and Hanslick concludes
that we cannot identify the beauty in music with the exactitude of representation
of feelings. They are separate principles, since musical beauty demands an
autonomous content, whereas the representation subordinates the beauty to
non-musical factors.35 The principle of autonomous beauty and the principle of
representation any non-musical content are, therefore, two separate principles.
In addition, in recitative we have separately the principle of musical beauty and
the declamatory principle (adjusting the music to sound like speech), in opera
the well-known struggle between between the principle of dramatic realism and
that of musical beauty, just as in dance and ballet there is a struggle between
the dramatic principle and the principle of plastic and rhythmic beauty.36 In all
these art forms, we can see different principles and each of them has its own
demands, which go in different directions. By this Hanslick tries to sort out the
specifically musical (or formal) principle from all the other principles we might
equate it with. The musical principle is directed towards forming autonomous
musical beauty.
Hanslick further tries to explain what beauty in music really is and what it
consists of if not in representation or arousal of feelings. This constitutes the so-
called positive thesis.
Formalistic definition of the beauty in music

Hanslick defines the beauty in music as specifically musical kind of beauty:
beauty which, irrespective of some non-musical content, consists solely in
tones and their artistic combination.37 Hanslick’s formalism is apparent in that
he considers the content of music to be the form of music itself. In addition,
he bases the criterion of beauty on the form of music, which I will elaborate
bellow. Namely, the material (Material) out of which the composer creates – the
substance of music – is tones, with their latent possibilities for melodic, harmonic
and rhythmic variety. The content, i.e. what is expressed by means of this tone-
material, is musical ideas, or tone-ideas, not some non-musical ideas concerning,
say, feelings, which would then have to be translated to the tone-language.38 The
form, consisting of tone structures of the composition, is a realized artistic idea,
or the embodiment of the content. The relation between content and form is not,
34 See OMB, p. 26, OML, p. 81. At this point Hanslick speaks of representation and expression
interchangeably, because his focus is on showing the difference between musical and any
other non-musical principle (which involves both representation and expression of feelings).
35 See OMB, p. 22, OML, p. 76.
36 See OMB, p. 22–3, OML, p. 77–8.
37 See OMB, p. 27–8, OML, p. 81–3.
38 See OMB, p. 28, OML, p. 83–4.
as is typically thought, the translation of an idea into some medium disparate

from it. It is, rather, a relation of gradual elaboration, detailed development to
the final form.39 Since tone-ideas – ideal tone structures – are the content of
music, the content of a composition is inseparable from its form (because the
form also consists in tone structures).40 In other words, the tone structures we
hear in a composition express just ideas of those same tone structures, because,
as Hanslick says, „music speaks not merely by means of tones, it speaks only
tones.”41 All this is implicitly contained in Hanslick’s definition of the content of
music as „tonally moving forms.”42 I would add that in this definition Hanslick
also succinctly included the time dimension of music: music forms do not stand
motionless, they move through time by the successive sounding of tones they
are made of, or, in Hanslick’s words, they come into being „in continuous self-
formation before our eyes.”43 This tone motion does not convey any other, non-
musical content as is ordinarily believed, and the beauty in music is the beauty
of these tone-forms themselves. In order to make his point clearer, Hanslick
gives examples of such purely formal beauty, beauty with no content from the
outside: arabesque, kaleidoscope, and later he adds works of architecture, dance,
human body and natural beauty such as a landscape, a leaf or a flower. All these
examples, Hanslick notes, also possess some kind of primitive beauty of outline
and colour, setting outside any expression or non-musical content. Hence, formal
aspects of all of them rest on the same basis, but unlike others, with the exception
of architecture works, musical works are produced by artistic endeavors and
express the ideas of the artist.44 For that reason, we have a different relation to
formal beauty in music then to the beauty of non-artistic objects. This leads us
to, in my opinion, the key point of Hanslick’s view on music and musical beauty,
which is completely neglected in many, especially contemporary interpretations.
Namely, musical beauty is, for Hanslick, defined dually: first, concerning only
our senses, as acoustical beauty in analogy with just mentioned visual examples,
39 See OMB, p. 28, OML, p. 83–4: “A musical idea turns up in the rough in the composer’s
imagination; it takes shape progressively, like a crystal, until imperceptibly the form of the
completed product stands before him in its main outlines.”
40 See OMB, p. 80, OML, p. 171: “But in music we see content and form, material and
configuration, image and idea, fused in an obscure, inseparable unity.” and “In music there is
no content as opposed to form, because music has no form other than the content.”
41 See OMB, p. 8, OML, p. 55, and OMB, p. 78, OML, p. 167. Also: Music says everything the
composer wanted to say – see OMB, p. 37–8, OML, p. 100.
42 „Der Inhalt der Musik sind tönend bewegte Formen“ – see OMB, p. 29, 101–2, OML, p.
84. The literal translation would be that the content of music are soundingly moving forms.
However, it should be noted that Hanslick consistently throughout the treatise uses the word
“ton” in the sense of tone, and not sound (for this he uses the term “Klang”), see OMB, p.
102. Therefore, soundingly is not an entirely adequate translation of Hanslick’s tönend, since
it omits the difference between a tone and a sound. Alternatively, the content of music could
be defined as “forms moved by the sounding of tones,” or, simply, as “tonally moving forms,”
which makes Payzant’s final choice.
43 OMB, p. 29, OML, p. 84: “(...) in fortwährender Selbstbildung vor unsern Augen entstehend.”
44 See OMB, p. 29, OML, p. 84–6.
and, second, concerning the artistic aspect and the art of shaping tone material,
as artistic value of embodied musical ideas, or externalized tone forms.
The first component – sensuous beauty of tone forms – is determined by
certain fundamental laws of nature, which govern the external manifestations of
sound and interrelations of all musical elements, and also the human organism
(e.g. which frequencies of sound we hear as pleasant, what we hear as harmony
and what as disharmony of tones etc.). These (negative, as Hanslick names them)
regularities, which are inherent in the tonal system due to the laws of nature,
are the basis for further capacity of tones for entering into the so-called positive
content of beauty, which in fact constitutes the artistic aspect of musical beauty.45
The realization of artistic ideas is achieved not through linking tones
mechanically into a series, but by spontaneous work of the composer upon the
tone material, the artistic structuring of tone relationships. The artistic value of
the ideas thus realized is the second component of musical beauty. Hanslick does
not give details about what exactly makes a composition artistically successful,
but only emphasizes that it always concerns the form of the composition.
Artistic ideas and artistic treatment of the material are immanent in the form:
tone structures we hear in music are realized artistic ideas themselves, and in
listening to the tones, we are able to follow the artistic acts of the composer,
which can be successful or unsuccessful, and thus we can evaluate them in terms
of their artistic value.46
Musical beauty, therefore, consists in the artistic value and acoustic beauty
of a formed (or, realized) tone idea.47 This dual definition of beauty is reflected
also in Hanslick’s claim that we notice musical beauty (it „makes itself known in
aesthetical awareness”) immediately, by itself, through the harmony of its parts,
and without reference to any factor external to the harmonious form of music,48
but, unlike contemplating the beauty of non-artistic objects, in listening to music
we do not stay on the surface of its sensuous aspect. Artistic beauty is designed by a
creative mind, and we, thus, in the structure of the composition and in the beauty
of that structure, consciously follow the ideas, or the artistic acts of its creator.49
45 See OMB, p. 30–1, OML, p. 88. Hanslick already noted that specifically musical beauty must
not be understood as simply acoustic beauty (which musical beauty often contains, but
cannot be reduced to), nor as an ear-pleasing play of tones, see OMB, p. 30, OML, p. 87. It
is clear that he does not take the term beautiful in the usual sense, and that by it Hanslick
understands not only the sensuous, but also the artistic aspect of music.
46 See OMB, p. 31, OML, p. 88–9.
47 Hanslick’s including of acoustical pleasure into his understanding of beauty is also evident
when he cites Grillparzer’s comments on the ugly, or unbeautiful, where he claims that, as
opposed to poetry, music cannot make use of the ugly in its works, since the impression of
music is received and enjoyed directly from sensation, so the understanding’s approval comes
too late to compensate for the intrusion of the ugly. “Hence Shakespeare can go all the way to
the hideous, but Mozart has to stay within the limits of the beautiful.” – See fn 18, OMB, p.
31, OML, p. 186.
48 See OMB, p. 32, OML, p. 90.
49 See OMB, p. 60, OML, p. 133–4: The second component, or the „ideal content,“ as
Hanslick says, can be understood only by the cultivated listeners, while the first component
is characteristic for the pleasure of the naive audience that exists in every art. Hanslick
Describing music in terms relating to emotions

The formalistic definition exposed in previous passage leads us to the next
significant point of Hanslick’s view on music. Namely, when we speak about
music, we describe it in various ways, but mostly by using terms that relate to
non-musical phenomena, especially feelings. Hanslick was not only well aware of
this, but was often describing music in that way himself. In addition, he explained
why we have the tendency to use these terms when describing music, and why
this phenomenon does not have the significance we are inclined to ascribe to it.
Firstly, Hanslick claims that the content of music cannot be expressed in
words or subsumed under concepts: music has sense and logic, but musical
sense and musical logic. Music is „a kind of language we speak and understand
yet cannot translate.”50 Therefore, it is extraordinarily difficult to describe this
specifically musical: purely musical content and specifically musical beauty.
Since music has no prototype in nature to imitate, and expresses no conceptual
content, Hanslick claims that it can be talked about only in technical definitions
or figuratively – „What in every other art is still description is in music already
metaphor.”51 Thus we say, for example, that one particular musical conception
is, taken by itself, witty, while another is banal, or sounds impressive, or insipid.
Hanslick emphasizes that we quite rightly describe a musical theme as majestic,
graceful, tender, dull, hackneyed, but all these expressions describe the musical
character of the theme.52 This means that by describing music in such ways, we
are not talking about the feelings of the composer, or the feelings that would be
the content of the theme. We are talking about the impression the combination
of musical factors used in that theme makes.53 In order to describe musical
characters, we often choose terms from the vocabulary of our emotional life,
such as arrogant, peevish, tender, spirited, yearning. However, we can also take
our descriptions from other realms of phenomena and speak of fragrant, vernal,
hazy, chilly music. Feelings are thus, for the description of musical characters,
only one source among others which offer similarities (Ähnlichkeit).54 Thus,
consideres both of these approaches as legitimate, aesthetical approaches (since the listeners
direct their attention on the artwork and its details), as opposed to the so-called pathological
reception which involves a passive enjoyment only in the general impression of the piece and
in the feelings wafting through it.
50 See OMB, p. 30, OML, p. 87.
51 OMB, p. 30, OML, p. 86.
52 See OMB, p. 32, OML, p. 90.
53 See OMB, p. 32–3, OML, p. 91–2. If the impressions and descriptions of musical
compositions were uniform, then, even if the descriptions were merely figurative, it might
indicate an objective connection between the tone groups and non-musical phenomena.
However, Hanslick has already shown that the impression of the composition (and hence the
description as well) in some way or another is, even in the same listener dependent on the
previous musical experiences, on non-musical associations (titles, programs, etc.), as well as
on various other factors. Therefore, describing music in the mentioned way by all odds is not
uniform, but to a great extent diverse and variable.
54 See OMB, p. 32, OML, p. 90–1. It seems that this passage is somewhat ambiguous since Hanslick
does not explain what he means by the term similarities. Budd is inclined to understand it as
Hanslick believes that describing music in terms relating to non-musical

phenomena does not imply the existence of some objective relation between
music and those phenomena, whether it is feelings, weather conditions, or
phenomena of some other realm. It is proper to use such descriptions, which
we even cannot do without, provided we never lose sight of of the fact that we
are using them only figuratively. We may say, for instance, that a musical theme
sounds gloomy, but we should take care not to say it portrays arrogance, or that it
is an expression of the gloomy feelings in the composer etc.55 Such descriptions
indicate the effect a composition has due to its musical factors, or the combination
of musical elements in it. In addition, Hanslick notes that since the individual
musical elements already possess their own characteristic expressiveness, the
predominant characteristics of the composer, such as sentimentality, energy,
serenity etc. will be revealed through his „partiality toward certain tonalities,
rhythms, transitions, in accordance with the prevailing impulse which the music
is able to reproduce.”56 However, Hanslick stresses in another passage that,
although there is, of course, a connection between the character of every piece of
music and the character of its author, this connection, firstly, in not visible to the
aesthetician, and, secondly, the comparison of the characters of musical pieces
and their authors is infinitely complicated and will be more prone to fallacies,
the stricter the causal connection it seeks to establish.57
I consider it obvious, especially in the last passage, that Hanslick did not
support the radical position which is usually ascribed to him. Moreover, his
view on the relation between music and emotions is compatible with the views
of contemporary authors who claim to disagree with him. I believe that the
key difference between Hanslick’s and their views is actually the difference in
emphasis. Namely, in the treatise On the musically beautiful, Hanslick’s main
intention was to refute the claims of those who were, in his opinion, actively
dynamic similarity since dynamic is the only aspect that Hanslick says is common for music
and feelings – see Budd, op. cit, p. 31. However, I wish to offer an explanation according to
which Hanslick does not presuppose a real similarity between music and other phenomena:
a common aspect such as dynamics, which enables us to describe music by using words
in their literal meaning (rapid, slowly, etc.). Hanslick himself says that when we describe
music, we mostly use figurative speech. I, however, take this figurative use to be a use of the
words in what Wittgenstein would call their secondary meaning – see Ludwig Wittgentstein,
Philosophical investigations, Basil Blackwell Ltd 1958, p. 216. This kind of use is exemplified
when we use the term “warm” to describe sounds (e.g. the sound of flute), colours (yellow),
even smells (the smell of vanilla). The term “warm” is clearly not used in its basic – literal –
meaning, but we also would not be inclined to say this is a metaphor which could somehow be
unraveled: what we are trying to say by using these kinds of descriptions we cannot say in any
other way but with that exact term. Likewise, music cannot literally be melancholic, fragrant,
or chilly, but neither can those descriptions be reduced to the descriptions containing some
other terms with no loss of sense. This is consistent with Hanslick’s statement in the same
passage that we cannot do without these descriptions. I believe that this is the way Hanslick
implicitly understands descriptions of music by using terms concerning emotions, as well as
weather conditions and other kinds of phenomena.
55 See OMB, p. 32, OML, p. 91, and OMB, p. 47, OML, p. 113.
56 See OMB, p. 47, OML, p. 112.
57 See OMB, p. 39, OML, p. 102
misleading the listeners, who then would not be able to recognize the most
essential aspect of music – the music itself, or the art of shaping tone material.
Hence, Hanslick displaces emotions aside: we should focus on music, and not
the emotions, regardless of the way they relate to music. Contemporary authors,
on the other hand, emphasize the relation between music (regardless of how
it is defined) and the emotions, and consider the importance of emotions in
our reception of music. Contemporary authors do not notice that unspecific
emotional expressiveness of music is indeed compatible with the main intention
of Hanslick’s treatise. Here I shall end my interpretation of Hanslick’s view.
Further, I will expose the interpretations of the same treatise, made by several
most influential contemporary music aestheticians, and then try to demonstrate
why those interpretations should be refuted.
Contemporary interpretations of Hanslick’s view

Contemporary authors mostly write about the negative part of Hanslick’s
treatise, and interpret only his arguments against the so-called emotional
aesthetics. I believe that their reconstruction of Hanslick’s arguments and
conclusions is incorrect, in interpreting both the arguments alone, as well as the
relations between the arguments and the conclusions. I shall not go into every
detail of their interpretations, but, rather, try to prove my thesis on several points
I consider most relevant.
According to Malcolm Budd, Hanslick tries to establish three negative
conclusions about the relationship between music and the emotions. The first
one states that it is impossible for any definite emotion to be represented by a
piece of music, and, hence, that the musical value, the „beauty”, of a piece of
music is never dependent on representing emotions by music.58
Budd only presents one of many Hanslick’s arguments for this conclusion –
the conceptual content argument. Budd reconstructs this argument as following:
music cannot represent thoughts, and definite emotions contain or involve
thoughts; therefore, music cannot represent definite emotions.59 Budd mentions
that, in this argument, Hanslick stresses that definite emotions are distinguished
from each other only by the thoughts they contain, that every other feature of
an emotion can be shared with others. However, it seems that Budd takes what
is contained in the emotion as the key point of the argument.60 Accordingly,
he takes the argument to depend on the theory of emotions we adopt, on how
we understand the notion of representation, on the idea of one phenomenon
containing or involving the other, and also upon the validity of the principle
that if one thing involves another thing then in order for something to represent
the first thing it must represent the second thing.61 I believe that, by putting
58 See Budd, op. cit, p. 20 Budd takes Hanslick’s musically beautiful to be equivalent to the value
of a musical piece.
59 Ibid, p. 21.
60 Ibid, p. 22.
61 Ibid, p. 21–4.
emphasis on what emotions contain instead on how they are distinguished

from each other, Budd made this argument more complicated than it is, and
rendered it dependent upon the considerations upon which, it seems to me, it
does not depend. Contemporary authors use a different view on emotions to
try to examine or refute this argument. Hanslick, however, does not engage in
any serious theoretical consideration concerning what emotions are exactly,
what they involve etc. He supposes, on an intuitive level, what may be specific
for particular emotions, and how they can be distinguished from each other
(among other, by the listeners whom music would represent emotions to), and
what of all that can music represent, or imitate. Although it may seem that
this argument would have been more convincing if it had been grounded on
a theory of emotions, I believe that its strength actually lies in the fact that it is
formulated in this way, more on an intuitive than on a theoretical level. I shall
try to demonstrate this point further in the paper, in the exposition of Jenefer
Robinson’s interpretation of the same argument.
Budd then presents three arguments for the first main conclusion – that
the value of a composition would not depend upon their being accurate
representations of definite emotions, even if such representation were possible.62
The first argument is that some valuable pieces of music do not represent a
definite feeling. Budd’s objection to this argument is that it counts only against
the stronger view that each valuable work represents definite emotions and its
musical value is in some way dependent upon this function. It leaves untouched
the view that the value of some music pieces is a matter of music’s representing
definite emotions. However, it is Budd’s formulation of the conclusion that
changes the point of the argument, and renders it unsuccessful. Hanslick does
not, in fact, speak about the relation of dependence, but about a relation of
congruence between musical beauty and the representation of feelings, and this is
supported by the text of his treatise.63 When the conclusion is formulated so that
it states that musical beauty and the exactitude of the representation of feelings
do not coincide, and that the former does not consist in the latter, the argument
becomes successful.
The second argument concerns the alleged inverse proportion between the
beauty of music and the exactitude of the representation of feelings by music.
In Budd’s version, Hanslick claims that vocal music, as it represents feelings
more perfectly (in the sense of succeeding to represent them), becomes musically
less beautiful. Since this turns to contradict Hanslick’s own belief that such
representation is impossible, Budd tries to save the argument by interpreting
62 Ibid, p. 25.
63 See OMB, p. 21, OML, p. 74: “Even if it were possible for feelings to be represented by music,
the degree of beauty in the music would not correspond (kongruieren) to the degree of
exactitude with which the music represented them. Let us for the moment, however, suppose
that it [the representation of feelings] is possible and consider the practical consequences.”
The arguments that follow are, in fact, designed to show the incongruity of musical beauty
with the representation of feelings, not the independence from it. Budd, on the other hand,
does not leave a reference for his formulation of the conclusion, but it is clear that he refers to
the same passage.
Hanslick as saying that as the words represent definite feelings more accurately, so
the music becomes less beautiful.64 This formulation does indeed avoid creating
a contradiction, but, unfortunately, does not make much sense either. Budd
actually does not notice that Hanslick does not speak about the degree to which
music succeeds in representing a feeling, but rather about the degree to which
music was tailored to representation, or adjusted so that it fits some non-musical
idea, and how much it was, on the other hand, composed only with musical
beauty in mind. I consider this to be inverse proportion Hanslick mentions. On
top of that, this was not even an argument that was supposed to prove anything,
but rather a passing comment about vocal pieces that were mostly dedicated to
the musical portrayal of feelings.65
The last argument for the same conclusion concerns the slight alterations
that can be made to the music in any song, which do not affect the accuracy of
the representation of feelings at all, but which destroy the beauty of the theme.
According to Budd, all that Hanslick intends is merely that the reference made
by the words of the song to the particular emotions will always remain, while the
overall beauty of the piece is affected. This does not succeed in supporting what
Budd takes to be Hanslick’s thesis: that the value of a composition never depends
on the representation of emotions by music.66 In my interpretation, on the other
hand, this actually is a successful argument, since it shows that it is possible that
the accuracy of musical expression of the content of the text – within the limits
of representing the dynamic aspect – remains the same, while the beauty of the
music is changed. This does support what I take to be Hanslick’s conclusion:
that musical beauty of a composition and the accuracy of representation (or,
as in this case, expression) are not the same thing. This interpretation not only
renders Hanslick’s argumentation far more sensible, but also accords more with
other Hanslick’s claims.67
Hanslick’s second conclusion concerns the arousal of emotions. According
to Budd’s interpretation, Hanslick here rejects the arousal of emotions as an
aesthetically relevant relation between music and emotions. Budd’s objection is
that Hanslick fails to recognize the distinction between the so-called musical and
extra-musical emotions, that is, the emotions which have the music as their object
(e.g. being impressed by a composition), and emotions which have an object
64 See Budd, op. cit, p. 26.

65 See OMB, p. 22, OML, p. 76: “(...) there are vocal compositions which try to portray a specific
feeling with the greatest accuracy, within the limits we have just explained, and in which
truth of portrayal has precedence over any other principle. Upon closer examination, the
outcome is that even the most relentless fitting of music to feeling in such a musical portrait
generally succeeds in inverse proportion to the autonomous beauty of the music (...)”
66 See Budd, op. cit, p. 26–7.
67 See OMB, p. 21–22, OML, p. 75. When discussing this very group of arguments, Hanslick
mentions that “music has the power to animate the object, to comment, and to bestow upon it
in greater or lesser degree the expression of individual subjectivity. It does this as far as possible
through the characteristics of motion and through exploitation of the inherently symbolic
aspects of tones.” Hanslick’s counterfactual approach in these arguments consists in supposing
that dynamic representation is sufficient for musical representation of a definite feeling.
aside from music (e.g. being sad over some event while listening to music). Budd
considers it unnecessary to deny, as Hanslick does, that an extra-musical emotion
can be experienced in an aesthetic response to music in order to maintain the
doctrine of the specifically musical nature of the beauty of music.68 I believe
that here also Budd misses some Hanslick’s statements which clearly show that
he did recognize the existence of the so-called musical emotions.69 This is most
obvious in the chapter in which he explains the difference between an aesthetic
and non-aesthetic reception of music. Hanslick rejects as aesthetically irrelevant
only those emotions that are aroused in passive listening to music, during which
we do not focus on the music itself.70
Budd believes that the third way it might be thought that music can be
related to an emotion in an aesthetically significant manner is by music’s being
expressive of an emotion or by music’s possessing an emotional quality.71
Denying that describing music as proud, longing, etc. implies an aesthetically
relevant relation between music and the emotions is, according to Budd,
Hanslick’s third negative conclusion. Hanslick allegedly tries to defend this
conclusion by stating that descriptions of music that contain emotional terms are
either improper or deletable.72 Budd then explains why Hanslick is mistaken to
claim that emotional terms are deletable when they are used this way,73 and on
top of that stresses that it is unclear how this figurative use of words is supposed
to work, for the only analogy Hanslick allows between an emotion and a piece of
music is their dynamic features, and yet neither feature that Hanslick mentions
as an example is not a dynamic feature (longing, fresh, etc.).74 I take that the
mistake in Budd’s interpretation is in misunderstanding Hanslick’s statement
that, when we describe music, just as we use the terms concerning emotions,
68 See Budd, op. cit, p. 27, 30–1.
69 For example: “Far be it from us to want to underestimate the authority of feeling over music.
But this [aroused] feeling, which in fact to a greater or lesser degree unites itself with pure
contemplation, can only be regarded as artistic when it remains aware of its aesthetic origin,
i.e., the pleasure in just this one particular beauty. If this awareness is lacking, if there is no free
contemplation of the specifically musical beauty, and if feeling thinks of itself as only involved
in the natural power of tones, then, the more vigorously the impression makes its appearance,
all the less can art ascribe such impression to itself.” See OMB, p. 58, OML, p. 131–2.
70 According to Hanslick, the aesthetical significance of a response to music does not depend
on whether emotions are aroused, but rather on whether it involves active and aware
observing, perceiving of the specificities of the work, the awareness of the beauty of the work
and listening to the piece only for its own sake, or, on the contrary, passive enjoyment in the
general character of the piece, without noticing its specificities, as well as listening to the
piece for the sake of enjoying one’s own emotions that do not take music as their object –
OMB, p. 63–6, OML, p. 141–5.
71 See Budd, op. cit, p. 31.
72 Ibid, p. 32. By improper descriptions Budd means that their use involves an essential
reference to the emotions, and by deletable that they can be replaced by sets of terms drawn
from different realms, which would serve exactly the same purpose. These other sets are in
turn deletable in favour of purely musical, that is, literal, characterizations.
73 Ibid, p. 31–5.
74 Ibid, p. 31.
„we can also take our descriptions from other realms of appearance,” as meaning
that there is no special reason why, say, „sad” should be used to stand for some
property, since the term is no more appropriate than many other words.75 I
interpret the same sentence to mean that we generally describe music by using
terms concerning various realms of phenomena, and so that emotional terms
are simply not the only set of terms we use to describe music. According to my
interpretation, Hanslick does not take the descriptions which contain emotional
terms to be deletable, but rather believes that some descriptions of music can
contain terms concerning emotions, while others can contain terms concerning
weather conditions or some other realms of phenomena, and that descriptions
themselves do not imply some special relation between music and those
phenomena in any of the realms. Therefore, since emotional terms are not the
only set of terms we use to describe music, there is no reason to believe that the
mere fact that we describe music using emotional terms could imply that music is
in some aesthetically relevant relation only with emotions. My interpretation, as
opposed to Budd’s, not only renders this passage coherent with other Hanslick’s
statements, but also has more support in the text of the treatise.76
Another significant intepreter of Hanslick’s statements is Peter Kivy, who
considers Hanslick’s intentions so unclear that he dedicated a whole paper to
figuring out what Hanslick even tried to deny in his treatise. Kivy’s interpretation
is supposed, among other things, to patch up many apparent contradictions that
arise between the arguments of the negative thesis and the rest of the treatise.
Kivy states that there are three clearly discernible arguments that weave in and
out of Hanslick’s negative thesis (although, because of the vagaries of Hanslick’s
style, he allows the possibility of there being more arguments), and all three are
directed against the possibility of representation of emotions by music.77
The argument Kivy marks as the first one was, in my interpretation,
concerned with the justification for ascribing any particular emotional content
to an artwork. Kivy, however, names it „the argument from disagreement” and
presents it as follows: if music could represent emotions, there would be general
agreement on what, in any given instance, a piece or passage of music represents.
However, since, according to Hanslick, there is complete, chaotic disagreement,
music cannot sensibly be thought to represent emotions.78 Kivy refutes this
argument by claiming there is no such disagreement. Besides oversimplifying
this argument, Kivy fails to notice the already mentioned consequence of some
possible disagreement: that if any two listeners disagree on which emotion is
represented by a composition, the music itself does not provide a criterion
75 Ibid, p. 33.
76 See OMB, p. 32, OML, p. 90–1: “Feelings are thus, for the description of musical characteristics,
only one source among others which offer similarities. We may use such epithets to describe
music (indeed we cannot do without them), provided we never lose sight of the fact that we
are using them only figuratively and take care not to say such things as ‘This music portrays
arrogance,’ etc.”
77 See Peter Kivy, ‘What was Hanslick Denying’, The Journal of Musicology, Vol. 8, No. 1 (1990),
pp. 3–18, p. 6. (in further text WWHD)
78 Ibid, p. 7.
for determining who of them is correct. Because there is nothing in the tones
themselves that would unambiguously point out to some definite emotion (in
Hanslick’s words, „there is no necessary connection”), there is nothing we can
use to justify the decision. Hanslick does not claim there is a „complete, chaotic
disagreement” on what is the content of musical compositions, but rather that
the issue of music’s emotional content would be much simpler and more evident
if music could represent definite emotions. I believe that this version of the
argument has more strength than Kivy’s version, and also puts forward a much
more interesting observation than it would if it was merely noting a statistical
fact about the agreement or disagreement of the listeners. On top of all this, my
version is supported by Hanslick’s explicit mentioning of justifying and refuting
attribution of emotional content to musical compositions.79
The second argument concerns the re-use of music by the composers to set
the texts of disparate emotive content or tone. Kivy presents it as an argument
trying to demonstrate that, since in those cases the disparity between the old
music and the new text is not perceived, it follows that music must be emotively
„neutral”.80 Kivy takes Hanslick’s statement that we can imagine a dramatically
effective melody which is supposed to express anger, but might also just as
effectively suit words expressing the exact opposite, say, passionate love, to mean
that every vigorous melody well suited to passionate anger is always well suited
(on the same account) to passionate love.81 I believe that this interpretation
oversimplifies the argument, and presents the conclusion incorrectly. It turns out
that, in Kivy’s version, Hanslick claims something that is clearly false (that music
is emotively neutral) and contradicts his own descriptions of music, and, on top
of that, uses the examples that quite obviously do not prove his point. In addition,
there are passages in the treatise which evidently show that for Hanslick music
can be more or less expressive of some emotions, but even when it is expressive to
a great extent, it is still insufficient for the achievement of one definite emotion.82
According to my interpretation, Hanslick characterizes music as emotively
indefinite, rather than neutral. This is the conclusion that he (successfully, I
believe) supports with examples of the same music suiting texts with a different
emotional content. Moreover, my version of the argument is coherent with
the rest of the treatise. This passage is related to Hanslick’s statement that
Gluck could have chosen tones more specific for the expression of passionate
grief for Orpheus’ aria.83 Kivy dedicates another paper to the clarification of
79 See OMB, p. 12–14, OML, p. 63–65.

80 Kivy, op. cit, p. 8.
81 Ibid, p. 8–9. Kivy easily refutes this version of the argument by giving a counterexample –
a musical passage that would not suit the words expressing love as it suits the words that
express anger.
82 See OMB, p. 18, OML, p. 69: “(...) music certainly possesses far more specific tones for the
expression of passionate grief.” Also: “But even far more specific and expressive (bestimmtere
und ausdrucksvollere) passages of vocal music will, when separated from their texts, at best
only allow us to guess which feelings they express. They are like silhouettes whose originals
we cannot recognize without someone giving us a hint as to their identity.”
83 See OMB, p. 17–8, OML, p. 67–9.
this statement, since it does not fit his interpretation of Hanslick’s view on the
expressiveness of music.84 In this paper, Kivy tries in many ways, but with no
success, to fit this statement with the rest of his interpretation of Hanslick’s view,
and ends by concluding that Hanslick was simply being incoherent.85 I believe
that this reaffirms that Hanslick’s view is not only more coherent, but also easier
to understand in my interpretation.86
The third argument is the already presented argument concerning the
conceptual content of emotions. In Kivy’s version, this argument demonstrates
the impossibility of musical representation of emotions based on what
emotions are and how they are ordinarily aroused. After noting that a good
deal of Hanslick’s view on emotions is „murky,” Kivy supposes that Hanslick
probably wanted to say that, since music cannot represent in sufficient detail
the situations in which emotions are normally aroused, it cannot be thought
to represent emotions. To this Kivy objects that the two claims clearly are not
equivalent.87 Kivy’s interpretation not only renders this argument unclear and
quite bad, it is also not supported by the text of the treatise. When speaking
of the conceptual content, Hanslick does not mention the situations in which
emotions are aroused, but rather the conceptions and thoughts which appear in
association with the emotion, during the occurrence of the emotion, although
Hanslick adds that they can appear also unconsciously. Because of the latter, I
am inclined to take the conceptual content as also the conceptions and thoughts
we ordinarily associate with the concept of an emotion.88 Thus, according to
my interpretation, the conceptual content is not the situation which arouses the
emotion, nor is it necessarily in the consciousness of the person undergoing the
emotion, but rather it is something we are inclined to associate with the emotion,
or with the concept of that emotion. Jenefer Robinson, as well as Kivy, interprets
this argument with the emphasis on the theory of emotions, especially on the
84 See Peter Kivy, ‘Something I’ve Always Wanted to Know About Hanslick’, Journal of Aesthetics
and Art Criticism 46 (1988), p. 413–17.
85 Ibid, p. 416–7.
86 In order to show that this argument indeed succeeds in demonstrating the so-called emotive
indefiniteness, I want to add that Hanslick emphasizes that he selected this example out
of many, “first, because it concerns the composer – Gluck – to whom has been attributed
the greatest exactitude in dramatic expression and, second, because over the years so
many people have admired in this melody the feeling of intense grief which it expresses
in conjunction with those words.” It is clear from this that the melody which would much
more suit the expression of one emotion was evidently very successful for the expression of
another, completely different emotion – see OMB, p. 18, OML, p. 69.
87 Kivy, WWHD, p. 9–10.
88 Hanslick did not explicitly formulate the statement that conceptual content is not solely a
part in the experience of emotion, but also the content we are usually inclined to associate
with the concept of the emotion. However, I believe that this statement can still be ascribed
to him, since it renders his argumentation more coherent, which I will substantiate further in
the paper.
For the passage that goes against Kivy’s interpretation, see OMB, p. 9–10, OML, p. 57–8:
“Only on the basis of a number of ideas and judgements (perhaps unconsciously at moments
of strong feeling) can our state of mind congeal (verdichten) into this or that specific feeling.”
arousal of emotions: Hanslick anticipates the judgement theory of emotions,

according to which judgement is an essential component of the emotion, and the
emotions are caused by a judgement. Her version of Hanslick’s argument is that
music cannot represent definite feelings because it cannot represent the specific
representations or concepts that define particular emotions.89 Robinson believes
this argument is refuted solely because psychology and neuroscience reject the
theory of emotions she believe is the basis of this argument by showing that
the judgement occurs in the later part of the emotion process.90 I do not want
to engage in questioning how occurring of the components of emotions in a
different order affects the definition of emotions, but rather just restate my belief
that discussions between different psychological theories do not have a significant
impact on the validity of Hanslick’s argument. Hanslick does not specify what
exactly he believes emotions are, what they consist of, how they arise or proceed.
Robinson herself notices that Hanslick’s view on the music’s influence on out
moods can be compatible with the results of psychological research she relies
on, and which allegedly refute the cognitive theory of emotions.91 Therefore, I
believe that, whichever theory of emotions we choose, we distinguish between
the emotions primarily in the way Hanslick (in my interpretation) claims. This
makes refuting any background theory of emotions by itself insufficient for the
rejection of Hanslick’s argument.92 It should also be noted that this is the only
argument Robinson presents, and the only one other authors take to posses
certain strength as well.
Steven Davies too mentions it as the only somewhat convincing argument.93
In his detailed review of the literature on music concerning musical meaning
and the expression of emotions by music, Davies presents other Hanslick’s
arguments as well, but this is consisted almost solely in combining Budd’s and
Kivy’s interpretations.94 Since Davies does not give any significant original
contribution to the reconstruction of Hanslick’s argumentation, I do not find it
necessary to present parts of his review, but I think it should be noted that Davies’
89 See Jenefer Robinson, Deeper than Reason, Oxford NY, 2005, str. 295.
90 Ibid, p. 3, 7–8. These surveys have, according to Robinson, showed that emotions are
processes in which a special kind of affective appraisal induces characteristic physiological
and behavioral changes, and only afterwards is this succeeded by the so-called “cognitive
monitoring” of the situation.
91 Ibid, p. 462, fn 7.
92 Therefore, instead of redefining the emotions (since this may leave the assumptions of the
argument untouched), one should rather try to refute this argument by showing there is
something besides the conceptual content that can be determinant for a particular emotion
(that music can convey), that is, by finding some other criterion for distinguishing emotions.
93 See Davies, op. cit, p. 209.
94 Ibid, see p. 152–3. for the argument on eliminability of the descriptions of music involving
emotion related terms (taken over from Budd); p. 203–208. for the argument on the relation
between music and text (taken from Budd and Kivy); 209–221. for the main argument
about the conceptual content (taken from Budd); p. 246–9. for the so-called argument from
disagreement (taken from Kivy); p. 281–283. and 289–291. for the interpretation of Hanslick’s
view on the arousal of emotions (taken from Kivy).
book at least indicates the acceptance and the influence of the interpretations
presented in this paper.
Conclusion
I hope, first of all, that in this paper I succeeded in presenting a clear
reconstruction of Hanslick’s view on music – a reconstruction that renders his
statements comprehensible and coherent, and thus in presenting Hanslick as a far
better philosopher than contemporary interpreters give him credit for. Moreover,
this would justify his role in the history of aesthetics in terms of philosophical
significance, and not solely due to the fact that no one has hitherto had a serious
approach to the aesthetics of music.
Due to the space limit, I was not able to expose all the details of contemporary
author’s interpretations of Hanslick’s view. Nevertheless, I consider the material
presented in this paper substantiated and sufficient for showing that it is not
the fact that these interpretations contain some minor misunderstandings, but
it is rather the whole argumentation that is misinterpreted. On the one hand,
this renders their rejection of Hanslick’s argumentation understandable, but, on
the other hand, it indicates the possible need for a reconsideration of the same
arguments, that is, the version of the arguments in which they are substantiated
and coherent with each other.
I hope as well that I succeeded in the main intention of this paper, namely,
to demonstrate that contemporary aesthetics of music (more precisely, the part
of it that deals with the relation between music and the emotions) rests on the
misinterpretation of Hanslick’s statements, which should have an impact on the
alleged refutations of those statements, and also on the soundness of the theories
that count on the incorrectness of Hanslick’s argumentation.
References:
Malcolm Budd, Music and the Emotions: The Philosophical Theories, London, 1985.
Eduard Hanslick, Vom Musikalisch-Schönen, 13–15. Auflage, Leipzig, 1922.
Eduard Hanslick, On the Musically Beautiful, trans. Geoffrey Payzant,
Indianapolis, 1986.
Eduard Hanslik, O muzički lijepom, pr. Ivan Foht, BIGZ, 1977.
Stephen Davies, Musical Meaning and Expression, Ithaca NY, 1994.
Peter Kivy, ‘Something I’ve Always Wanted to Know About Hanslick’, Journal of
Aesthetics and Art Criticism 46 (1988), 413–17.
Peter Kivy, ‘What was Hanslick Denying’, The Journal of Musicology, Vol. 8, No.
1 (1990), pp. 3–18.
Jenefer Robinson, Deeper than Reason, Oxford NY, 2005.
Ludwig Wittgentstein, Philosophical investigations, Basil Blackwell Ltd 1958.
Ljiljana Radenović Original Scientific Paper
University of Belgrade UDK 165.242
165 Витгенштајн Л.
EMPIRICAL RESEARCH INTO THE PROBLEM

OF OTHER MINDS: HOW DEVELOPMENTAL
SOCIAL PSYCHOLOGY CAN STRENGTHEN THE
WITTGENSTEINIAN ATTITUDINAL APPROACH1
Abstract: My goal in this paper is to examine the Wittgensteinian attitudinal

approach to the problem of other minds initially proposed by M.R.M. ter Hark (1991)
and developed by Hyslop (1995). As some other Wittgensteinian approaches this one
also starts with the assumptions 1) that there is no asymmetry between self knowledge
and knowledge of others and 2) that the relation between our inner states and outer
behavior is a conceptual one. However, it differs from the standard Wittgensteinian
approaches, mainly in its claim that knowledge of others does not come in the form of
a belief that needs to be justified but rather in the form of an attitude toward others.
This attitude is an intuition that enables us to form beliefs about other people. To
explicate what this exactly means is my first goal in this paper. The second one is to
examine how viable this position is. I plan to do this in an unorthodox way and ask
psychologists to help me out. I argue that the insights we gain from developmental
psychology can cast some light on how children develop a ‘Wittgensteinian’ intuitive
attitude toward others. In the final section I turn to Hyslop’s claim that even this
approach which distances itself from beliefs and justification of beliefs cannot
overcome skeptical worries. Hyslop goes on to argue that our intuitions may be strong
but wrong as is the case with racist and sexist intuitions. My goal is to show that
Hyslop’s analogy between our attitude toward others and racist or sexist intuitions
fails and that a naturalized attitudinal approach, the way I present it, stands further
away from the skeptical attack than Hyslop has envisioned.
Key words: problem of other minds, attitudinal approach, skepticism, joint
attention, development
Even though it might appear that the problem of other minds raises the
question about the nature of our psychosocial strategies for making sense of
other people, their thoughts, feelings, and actions, this is not its focal point.
1 Ovaj članak nastao je u okviru projekta”Dinamički sistemi u prirodi i društvu: filozofski i

empirijski aspekti” (evidencioni broj 179041), koji finansira Ministarstvo prosvete, nauke i
tehnološkog razvoja Republike Srbije.
Philosophers have not typically been inquisitive about strategies that help us
realize, for instance, that a friend of ours is depressed or how often we get it right
when we speculate about what she thinks. The philosophical problem of other
minds runs deeper than this. This problem is about how I know if there were
other minds like mine at all and how I could justify such knowledge. In other
words, it is not simply about how I can know what other people think but about
how I can know if they think. This radical question that opens up an equally
radical gap between others and me is what is considered to be the traditional
philosophical problem of other minds. Formulated in this way the question of
other minds appears to be a textbook example of an epistemological question: it
is the question about the nature and justification of our knowledge.2 But, given
its radical nature, we might ask how did philosophers end up questioning the
existence of other minds on such fundamental level? Where and how does this
problem occur?
The philosophical problem of other minds is often described as a problem
that arises from the asymmetry between the way we get to know our own mental
states and the way we get to know that other people have similar states. While it
seems that we know ourselves directly, our knowledge of others (if we have such
knowledge) is indirect. The problem is often traced back to Descartes’ philosophy
while a variety of solutions have been proposed since then. Two major opposing
philosophical camps have emerged. On one hand, there are philosophers who
do accept this asymmetry of knowledge. Their goal is to show that despite the
fact that we cannot reach other people’s minds directly we do have enough
inductive evidence to know that they do have minds (see e.g. Mill 1872, Russell
1970, Hampshire 1970). On the other hand, philosophers working within the
Wittgensteinian tradition (such as e.g. Malcolm 1958, Hacker 1972) tend to
reject such asymmetry altogether arguing that the nature of our knowledge of
other minds as well as the evidence we have for it is not based on empirical
evidence at all. The crux of their argument is that the relationship between
our minds and our behavior is not causal and contingent (i.e. empirical in any
way) but conceptual. Thus, our knowledge of other people, their thoughts and
feelings, depends on our knowledge of the meaning of mental concepts such as
pain, joy and the like along with our estimate of whether people’s behavior fulfils
the criteria for being in pain or feeling happy. When these criteria are fulfilled
there is no space left for any doubt about the existence of their minds.
My goal in this paper is to examine one version of the Wittgensteinian
approach to the problem of other minds proposed by M.R.M. ter Hark (1991).
Hyslop (1995) refers to it as the Wittgensteinian attitudinal approach while making
cautionary note that this might not be the solution that Wittgenstein himself
had in mind. Nevertheless, the approach is Wittgensteinian as it starts with the
assumption that there is no asymmetry between self knowledge and knowledge of
others as well as that the relation between our inner states and outer behavior is a
2 Whether the problem is merely epistemological or comes with the whole package of
ontological assumptions has been debated on many occasions. For the review see e.g.
Avramides (2001).
Ljiljana Radenović: Empirical research into the problem of other minds... 137
conceptual one. However, it differs from the standard Wittgensteinian approaches,

mainly in its claim that knowledge of others does not come in the form of belief
that needs to be justified but rather in the form of the attitude toward others.
This attitude is an intuition that enables us to form beliefs in the first place. To
explicate what this exactly means is my first goal in this paper. The second one
is to examine how viable this position is. I plan to do this in an unorthodox
way and ask psychologists to help me out. I will argue that the insights we gain
from developmental psychology can cast some light on how children develop
a’Wittgensteinian’ intuitive attitude toward others. In a nutshell this will be my
attempt to naturalize the attitudinal approach to other minds. In the final section
I will turn to Hyslop’s claim that even this approach which distances itself from
beliefs and justification of beliefs cannot overcome skeptical worries. Hyslop goes
on to argue that our intuitions may be strong but wrong as is the case with racist
and sexist intuitions. My goal is to show that Hyslop’s analogy between our attitude
toward others and racist or sexist intuitions that we might develop about them
fails and that a naturalized attitudinal approach, the way I present it, stands further
away from the skeptical attack than Hyslop has envisioned.
1. The attitudinal approach

A variety of Wittgensteinian solutions3 to the problem of other minds
gained popularity among ordinary language philosophers of the twentieth
century. In its traditional form a Wittgensteinian approach is meant to be the
alternative to the analogical argument, and of course, unlike the analogical
argument, to be successful in answering the skeptic. So, let me briefly summarize
what the argument from analogy is all about. The argument is built on two
main presuppositions: the first one is epistemological and is about two kinds
of knowledge that we have while the second one is ontological and is about
the nature of the relationship between our inner mental states and our outer
behavior. So, let me take a closer look at both of these presuppositions.
According to J. S. Mill (1872), who was among first to explicitly formulate
the analogical argument, there is an important distinction between two kinds of
knowledge: knowledge that we get directly through consciousness (based on our
intuitions) and knowledge that we get through inference. Knowledge about our
own sensations comes to us directly while the knowledge about the outside world
including our knowledge of other minds must be inferred. But, how do we make
such inference about others? Mill argues that we initially notice the connection
between our own sensations and behavior. Having done this we acknowledge
that others behave in a similar fashion but that their behavior is not correlated
with our own sensations. From here we conclude that their behavior must be
connected to the sensations of their own. Now, Mill insists that the conclusion
3 There have been the old and the new Wittgensteinian arguments. The former has been
proposed and defended by Norman Malcolm (1958, 1959). The latter has been defended by
Saul Kripke (1982). As I am interested in the attitudinal approach proposed by M.R.M. ter
Hark I will not be concerned with the differences between the other two.
that we reach by analogy can be subsequently verified by further observations.

Mill thought that the repeated ongoing experiences and observations make this
inductive argument a good one.
At this point, we can ask what kind of implications Mill’s epistemology has
for the way we conceptualize the relation between our inner mental life and
our behavior. We can find the answer to this question in Russell’s version of the
argument (1970). Russell argues that there is a causal relation between our inner
mental states and outer behavior, a relation that we are able to observe directly.
Based on this observation we infer that a similar connection obtains in other
people. However, causal connection between my thoughts and feelings, on one
hand, and my behavior on the other does not obtain every time even in my own
case. Even if it did it would not be certain that it does apply to other people.
Thus, Russell concludes that causal connection when it comes to others cannot be
inferred with certainty. However, it can be inferred with high probability. To fight
the skeptic high probability is all that we need, though. Or so Russell thought.
The argument from analogy faces some serious problems. One of the main
worries with the argument is that our reasons for believing in other minds are
based on the sample of one, i.e. on observations of ourselves. As we know, a single
case is always a poor inductive base. Furthermore, we can use the main insights
from the famous Wittgenstein’s private language argument (1953) and conclude
from it that it is not even clear how we can intuitively detect the relationship
between our inner feelings and thoughts and outer behavior without publicly
shared language. This language enables us to identify and categorize these inner
feelings and thoughts in the first place. In other words, without publically shared
language we would not be able to intuitively know ourselves and relate our inner
life to outer behavior let alone apply this knowledge to the cases of others.
Some philosophers working in the Wittgensteinian tradition, such as
Malcolm (1958, 1959) or Hacker (1972), have argued that the argument from
analogy cannot possibly overcome these problems. So, instead of trying to fix it
we should reconsider the two aforementioned presuppositions, epistemological
as well as ontological, that the argument is built upon. This basically means that
we need to propose a new way of understanding our self knowledge as well as
the new way of conceptualizing the relationship between inner mental life and
outer behavior. So, how are we going to do this?
While writing Philosophical Investigations (1953) Wittgenstein was occupied
with the same worries. Even though there has been a lot of disagreement about
the proper interpretation of what Wittgenstein really had in mind and what
his position really was, most interpreters agree that he rejected the idea that
we know our own mental states by introspection and that the relation between
mental states and behavior is causal, thus rejecting both main presuppositions of
the analogical argument4. Let me elaborate both of these points.
4 An introductory survey of different interpretations of Wittgenstein’s take on language, mind,

and conceptual versus empirical connection between inner and outer can be found in: Miller,
Alexander and Crispin Wright, Rule-Following and Meaning (2002). However, I will focus on
the standard interpretation offered by Malcolm (1958, 1959) and Hacker (1972).
According to Malcom and Hacker Wittgenstein denied that we have

introspective knowledge about our inner states. This basically means that we
do not observe our own thoughts and feelings with some inner, intellectual eye.
Furthermore, we do not introspectively know, i.e. make the connection between
inner mental states and outer behavior. Indeed we are aware of how we feel,
what we think, and what we want to do. But, this self-awareness is not of an
introspective kind; the kind that presupposes that we are spectators of our own
mental life. Our awareness of how we feel or what we think does not constitute
theoretical or intuitive knowledge. It is neither indirect nor direct. In the same
time the relationship between our inner states and outer behavior is not external,
causal, and contingent. Our outer behavior is internally related to how we feel and
what we think and represents the expression of our inner life5. In other words,
inner mental life and behavior are not two different entities in a causal relationship
but rather two aspects of one phenomenon. I will say more about developmental
origins of this internal link between inner mental life and outer behavior, but
for now it is important to keep in mind that for Wittgenstein the crucial step in
rethinking the problem of other minds is to redefine the way we understand self-
knowledge and the way we conceptualize the inner/outer distinction.
Having abandoned the concept of introspective knowledge in our own case
and contingent empirical relations between our inner lives and behavior we can
proceed to apply these insights to the problem of other minds. So, assuming that
there is an internal link between our inner feelings and our behavior, such a link
needs to obtain in the case of other people as well. This means that, if such a link
does obtain, when we observe other people’s behavior we observe their mental
states too. But, what exactly does it mean ‘to observe the mental lives of others
in their behavior’ and how within Wittgensteinian position we can hope to avoid
skeptical worries with this move?
As we have seen according the Wittgensteinian view the relationship
between inner life and outer behavior is not empirical. The question is: if it is not
empirical what is it? Malcolm and Hacker are going to argue that the nature of
their relationship is conceptual. This means that our concept of behavior already
involves the concept of inner pains, joys, beliefs, desires and the like. We can make
an artificial distinction between the two (inner and outer) and think of them as
separate events or phenomena but ‘scream’ and ‘pain’ come together conceptually.
In our own case screaming is the expression of pain. In the case of others their
screaming meets the criteria by which we determine that they are in pain. More
precisely, this means that when we observe other people we judge whether their
behavior passes the criteria for what e.g. being in pain or being happy is. This is
what it means to observe pain in pain behavior. We do not conclude or infer from
scream that somebody is in pain. We see that they are in pain as we know what
pain means and that meaning involves a particular kind of behavior.
Now, this has important implication for the skeptical worries about other
minds. In the standard Wittgenstenian approach when somebody passes the
5 Even though Malcolm and Hacker portray Wittgenstein as holding firmly to this view there
are some indications pointed by L. Kojen (2009) that Wittgenstein himself had doubts that
all mental terms and their meanings could be identified and reduced to one function: our
expression of the inner mental life.
criteria for being in pain this serves as justification of our belief that they are
indeed in pain. However, this justification does not come from inductive
inference nor does it consist of gathering empirical evidence of any sort as the
relation between inner states and outer behavior is not causal but conceptual. So,
there is no need to worry about the inductive base that is too small nor about
the way we make intuitive connection between inner and outer (or the way we
generalize such knowledge to others) as we do not make such connection at all.
We know what other people feel because we know the meaning of the words
and since we know the meaning of the words we see their pain in their behavior.
Thus, when other people pass the criteria for being in pain then we can conclude
that they are indeed in pain without worrying about skeptical objections.
The Wittgensteinian approach I presented here is what is usually called
the criterial approach to other minds. Now, let me turn and see how it differs
from the other version of the Wittgensteinian approach, namely the attitudinal
approach. My plan is to clarify and defend the latter in the following sections.
While the criterial and the attitudinal approach share the assumptions related
to the inner/outer connection and the nature of self-knowledge the major
difference between the two lies in the way they conceptualize our knowledge
of others. What they have on common is as follows: a) the relationship between
our inner states and outer behavior is not external, causal, and contingent; b)
our awareness of how we feel or what we think does not constitute theoretical
or intuitive knowledge. It is neither indirect nor direct. However, within the
criterial approach our knowledge of others comes in the form of beliefs while
in the attitudinal approach the knowledge of others is not a set of beliefs at all6.
Given that the crietrial approach presupposes that we do indeed have beliefs
about other people’s minds (e.g. that somebody is in pain, or is happy, or believes
in gender equality) the question that the skeptic asks is whether we can justify
such beliefs? According to the criterial approach this justification does not come
from induction as the relation between inner life and behavior is not empirical.
However, justification does not and cannot come in the form of logical entailment
either as the relationship between the two is not strictly speaking logical. If the
connection were logical the inner feelings would entail outer manifestations. We
are all aware that this is not the case. People are able to hide or fake their pains.
There is no logical entailment between the two. As Hyslop correctly points out,
the standard criterial approach is meant to find a ‘midway between entailment
and induction’ to justify our knowledge of other minds. In a nutshell, the goal
of the criterial approach is to defend our knowledge about other minds within
the framework of standard folk psychology7. This basically means that such
approach presupposes that our knowledge of others comes in the form of beliefs.
6 As we will shortly see when I turn the developmental issues (development of mind, social
cognition, language) these two approaches are not necessarily mutually exclusive. However,
the research in developmental psychology indicates that the attitudinal approach to other
minds develops first and serves as the basis for language development within which the
cirterial approach to other minds makes sense.
7 ‘Folk psychology’ is the term used in many different ways in cognitive science and philosophy
of mind (for the review see e.g. Clarke 2001). However, when philosophers and cognitive
scientists talk about folk psychology they usually presuppose that folk psychology must
Now, if it comes in such form then it needs some sort of justification if it is to

avoid skeptical worries.
To summarize, according to many classical interpretations of Wittgenstein
(such as Malcolm’s and Hacker’s) Wittgenstein did argue for the conceptual
(not empirical) link between inner mental states and outer behavior while he
still seemed to be endorsing the view that our knowledge of others comes in
the form of beliefs and that our beliefs about other people mental states need
to be justified. According to these interpreters, we justify such beliefs through
examining the conceptual criteria for being in a certain mental state and whether
the person in question fulfils such criteria. When a person fulfils the criteria
for a particular mental state we are justified in believing that they are indeed in
such state. Malcolm’s famous example of scratching and itching nicely illustrates
this point (Malcolm 1959). We know that itching and scratching are related.
Within the criterial approach their relation is conceptual. So, when somebody
is scratching we know they are itching. This is not inductive or logical truth but
conceptual. Scratching fulfills the criteria for itching and we can justify our belief
that whoever is scratching that they are itching as well.
As Hyslop points out, this standard criterial approach faces a serious problem:
if the relationship between inner and outer is not the one of logical entailment
how could criterial justification provide certainty, (certainty that would satisfy the
skeptic) for our belief that other minds exist? Furthermore, given the possibility
of pretending and lying many would agree that psychobehavioural relations are a
bit further from logical entailment and closer to empirical contingency. To try to
be in between and say that their relationship is not empirical but of conceptual/
logical nature of a particular kind that does not involve logical entailment
remains unclear8.
Now, the question is can we find the way out of these worries but not fall
back to the already abandoned analogical arguments? It seems that there is still
one possible exit for those inclined to accept Wittgensteinian strategy but it is
the radical one. Namely, one possible solution is to argue that our knowledge
that other minds exist does not come in the form of belief at all and hence there
is no need to justify it. This requires some further unpacking.
First, if we banish the concept of knowledge, at least the concept of
knowledge that comes in the form of propositional attitudes (beliefs), when it
comes to mental states (either ours or that of others) the question is how we are
to think about the fact that we indeed seem to know how we feel and seem to
know (through observation) how other people feel. Hyslop tries to answer this
include human ability to form beliefs about beliefs of others, i.e. it must involve human ability
to ascribe beliefs to others in order to explain and predict their behavior.
8 I am grateful to the reviewer of this paper for the suggestion that the fulfillment of the criteria
could be ‘ceteris paribus’ or defeasible by other factors (such as the context). This would give us
something less than entailment but such fulfillment of the criteria would be categorically different
from induction. However, no matter how we clarify the status of the conceptual justification it
seems that the skeptical worry will not go away unless we have the certainty of logical entailment.
So, the question would still remain: can we overcome this worry and win against the skeptic even
if we successfully defend the independent status of conceptual justification?
by saying that: „Something perhaps deeper than knowledge applies in one’s own
case, something prelinguistic even.” (1995, p. 124). As for other people and their
minds Hyslop says that our knowledge of them is rather a form of an attitude
(also probably paralinguistic) that already involves treating others as minds
never as mere bodies. When Hyslop insists that we are aware of ourselves on a
deeper, prelinguistic level and that we have the approach to others that already
presupposes their minds he wants to differentiate linguistic knowledge that
usually comes in the form of propositions, propositional attitudes, shortly in the
form of beliefs from immediacy of self-awareness and our strong intuition that
other people indeed think and feel. The latter point is in a sense a step beyond
the Wittgensteinian point that inner life is conceptually related to outer behavior.
We should think of it as a prelinguistic intuition which tells us that others are
persons with their thoughts, feelings, and intentions. So, when we say that we
see mind in other people’s behavior this ‘seeing’ of the mind is happening on the
basic perceptual level and does not come out from some semantic knowledge
of the meaning of the words. It would follow then that the mental concepts
and their meanings as well as Wittgensteinian thesis that the relations between
behavior and inner life is conceptual have their roots in something more basic.
In other words, the attitudinal approach presupposes something stronger then
conceptual relation between inner and outer as it has roots in prelinguistic not
linguistic practices. Moreover, as we will see shortly these prelinguistic practices
are developmental prerequisite for normal language development. This means
that the conceptual relation between inner and outer emerges from paralinguistic
practices that we are engaged in. In the next section I will say more about this.
When we think about the phenomenology of our everyday experience
it seems that there is nothing problematic so far about the main theses of the
attitudinal approach. When we see someone fall down there is this immediacy
of being aware that they are in pain (even though we do not feel it) in the same
way there is the immediacy of feeling our own pain. There seems to be no in
between steps of introspection or inference in any of these cases. This immediacy
allows us to be certain about ourselves and others. According to the attitudinal
approach, this certainty does not come in the form of justified belief or belief at
all. But, the question still lingers: what is this attitudinal approach exactly and
how does such attitude give us such certainty? In the following section I will
examine whether we can clarify what this exactly means, what the attitudinal
approach involves, and what it implies. In order to do this I am going to use
some findings and insights that are coming from developmental science.
2. Development of social cognition:

some insights from psychology
Many philosophers involved in this discussion will find counter intuitive
the attitudinal approach described in the previous section. The main reason for
being wary of such an approach is because it seems somewhat uncontroversial
that our knowledge of other minds does indeed involve all sorts of beliefs and
that we have good or not so good reasons for holding such beliefs. To argue
that our knowledge of others does not even come in the form of beliefs seems
strange. Even if at the first glance such position appears to be promising in the
fight against skeptical worries it still seems to be raising more questions than it
promises to solve.
What is this attitude after all? What is its nature and how are we to think
about it? So far we only know what this attitude is not: it is not a belief and
it should be psychologically and epistemologically deeper than belief. It is
supposed to allow us to have the immediacy of knowing that others are in pain
without us having to do any inferring. However, in many cases it does seem that
we are engaged in a reasoning process trying to figure out how people really feel
and what they think. We use all sorts of strategies and all kinds of evidence to
come to the conclusion about their inner states. This kind of reasoning seems
to suggest that our knowledge of others does indeed come in the form of belief
and requires justification. The required justification could be only obtained
through induction or logical entailment. This brings us back to the dilemmas
and problems that both criterial and analogical approaches have faced.
Indeed, this brief and very general analysis of our everyday musings about
other people’s inner lives does not necessarily imply that we need to give up the
attitudinal approach altogether. However, in order to hold on to it we need to
specify more closely what exactly this attitudinal approach is. Furthermore, in
order to convince anybody to embrace such an approach we need to explain
why it seems that we do form beliefs about other people’s minds and why we
do search for evidence for the beliefs we have formed. That is, we need to show
how our everyday practices of forming and justifying beliefs emerge from the
attitudinal, prelinguistic knowledge.
In order to do this we need to step back from our everyday practices and
look at how such practices develop. More specifically, we need to take a look
at the development of social cognition. This is not something philosophers like
to do but it is nonetheless of crucial importance for this particular problem.
In the rest of this section I will try to outline the origins of this prelinguistic
attitude toward others as well as to show how our beliefs about others emerge
from prelinguistic knowledge that children develop throughout the first years
of life. I hope that this developmental journey will clarify in what way such
specific attitude toward others represents the foundation for more sophisticated
linguistic forms of social cognition.
Let us first see at what age children begin to understand that other people
have beliefs of their own and how people form such beliefs. Back in 1983 Heinz
Wimmer and Josef Perner developed a test that was meant to answer these
questions. This test is now known as false beliefs test. In the test children are
presented with the following story: Maxi places some chocolate in a cupboard in
the kitchen and leaves the room. While Maxi is away, another character takes the
chocolate from the first cupboard and puts it in the second cupboard, and then
leaves. When Maxi returns, the child is asked to predict where Maxi will look
for the chocolate. The correct answer, of course, is that Maxi will look for the
chocolate in the first cupboard because we know that Maxi could not know that
the chocolate was moved to another drawer. This experiment has shown that
children of four years and older do understand what Maxi can and cannot know
about this situation. Younger children do not have this understanding as yet and
cannot answer the questions about Maxi’s beliefs properly. Instead they tend to
apply what they know, i.e. their own beliefs, to Maxi.
However, even before children develop the ability to correctly attribute
beliefs to others they do engage in social interaction and show basic ability to
interpret other peoples’ intentions. It is not surprising that such ability starts very
early, during the first year of child’s life, even before language kicks in. I will trace
back and outline these prelinguistic origins of social cognition in the attempt
to show how this prelinguistic attitude toward others as intentional beings
(rather than mere objects) emerges. As we will see shortly the development of
social cognition in infants is the single most important predictor of successful
language acquisition. In the case that this development is derailed the language
development is derailed too. The case of children with autism illustrates well the
connection between prelinguistic social cognition and language development. I
will come back to this to this point at the end of this section.
The world in which the child is born is intrinsically social as the child needs
a caregiver if she is to survive. The caregiver needs to be there for the child not
only to feed her but to regulate her arousal states so that the child gets enough
sleep as well as enough external stimuli that are necessary for normal physical
and psychological development (see e.g. Shanker 2013). The child at birth does
not have the required regulatory systems in place so the caregiver serves as a
sort of the external regulating brain (Tantam 2009) to a child. If the child is to
be successful and develop quickly the child needs to be able to tune in and read
social cues coming from the caregiver. The better the child is in this non verbal
communication the better the development of her self-regulatory, cognitive,
affective systems will be. So, the first precursors of nonverbal communication
start at birth.
However, the most important step in development of such communication
is marked by the emergence of joint attention when the child is approximately
nine months old. Joint attention refers to the ability to „coordinate attention
between interactive social partners with respect to objects or events in order
to share awareness of the objects and events” (Mundy et al. 1986, p. 657). It is
important to note that the child who is able to engage in joint attention is able
to make use of the caregiver’s communicative signals. In other words, through
preverbal communicative signals, the child and the caregiver are able to follow
and direct each other’s attention. The clear sign that the child is able to make use
of the communicative signals of a caregiver can be seen in her ability to follow
the gaze of the caregiver and determine the object(s) to which the caregiver
intends her to attend.
It is very likely that the nature and the development of these preverbal
signals highly depend on the child’s tendency to attend to and make sense of
the caregiver’s emotional expressions. For one thing, being preverbal, these
signals consist of expressions of approval, disapproval, surprise, and the like

(i.e., they, at least partially, consist of emotional expressions). That is, along with
the more conventional communicative gestures, such as pointing9, emotional
expressions can efficiently direct the attention of a communicative partner to
a particular object or a property of an object in the joint attentional scene. For
another thing, joint attention episodes are frequently punctuated and motivated
by affective engagement with another (Paparella, D’Angiola, & Kasari 2001;
Paparella & Kasari 2002).
Joint attention as described here is apparently one of the crucial steps in
understanding other people’s minds. The child is not as yet able to comprehend
entirely what it means to have inner mental life and what exactly beliefs are
(nor is she able to differentiate her own beliefs from that of her mother) but is
nonetheless able to understand that other people want them to pay attention to
an object or want them to do something (or to stop doing something). Children
also become increasingly aware of their own agency, i.e. that they, themselves,
can also direct other people’s attention and make them do what they want. This
is all happening prior to language acquisition while communication is taking
place in the medium of gestures and emotional expressions. For the child the
expression on mother’s face is the emotion that mother feels and that emotion
tells the child how to proceed and what to do. The child is also able to direct
mother’s actions by using her own gestures and expressions. There is immediacy
in these co-regulated activities. Here we can find the origins of treating other
people as minds not as mere objects. Even before the child is able to engage
in joint attention the child does differentiate between inanimate and animate
objects (for review see Johnson 2000). With joint attention the child begins to
understand that other people have intentions. But the concept of belief and what
it means to have a belief and how these beliefs are formed is yet to be acquired.
The acquisition of such concept requires language10.
Now, the studies of children with autism have repeatedly shown that there
is a close connection between language acquisition and participation in the joint
attentional scenes. It is well known that the participation in joint attentional
scenes emerges (if at all) very late in the development of children with autism
(Charman 2003; Kasari, Sigman, Mundi, & Yirmiya 1990; Leekam, Lopez, &
Moore 2000; Loveland & Landry 1986; Mundy & Crowson 1997). Considerable
9 Given that the emergence of social orienting and attention to distress occurs long before the
child acquires conventional communicative gestures, it would be worthwhile to explore how
the former relate to the latter and whether the origins of communicative gestures lie in the
more basic understanding of emotional expressions.
10 It is not surprising that language acquisition is a crucial factor in category development,
concept acquisition, and the development of abstract thought. It has been shown that
language acquisition shapes the development of concept acquisition and category formation
(see. e.g. Markman 1989), and contributes in an important way to flexibility in categorization
tasks (see e.g. Ellis & Oakes 2006). In other words, once language kicks in, along with the
child’s ability to communicate linguistically, language is the main medium of social learning
and, as such, is the driving force behind category development and concept acquisition which
includes the acquisition of mental concepts such as pain and happiness and more abstract
mental concepts such as beliefs and desires.
delay in joint attention is followed by the delay of language acquisition and the
further impaired development of social cognition in these children. So, in a way
the prelinguistic understanding of others as minds with their own intentions is
the foundations of language acquisition and as such represents the single key to
the development of abstract thought that allows us to think about our own and
other people’s beliefs.
We can now pose a further question and ask what the origins of joint
attention are. In the literature, there are two general ways of explaining normal
development of joint attention. According to one approach, joint attention in
normally developing children is explained by emotional interaction with their
caregivers. As a result of normal emotional interaction they develop normal
intersubjective relations (Hobson 1993) and engage in normal socio-emotional
development (Mundy 1995). According to this approach, children with normal
socio-emotional development exhibit normal face-to-face gaze, direction of
attention, and appropriate patterning of behavior within interactions. All of the
above enables them to participate in joint attentional scenes. According to the
other approach, the origin of joint attention is primarily cognitive. Instead of
looking into socio-emotional development, the proponents of this view hold that
children develop their understanding of the caregiver’s attention to an object.
More precisely, children develop the ability to represent the relation between
caregiver and the object to which the caregiver is attending (Baron-Cohen 1995;
Baron-Cohen, Leslie, & Frith 1985).
However, stated in the most general terms, the two approaches provide only
a starting point for further research. Both views, if they are to be useful, need
further refinement. The goal of the subsequent research is to determine, for
example, in what way socio-emotional development brings about joint attention
and/or what underlying mechanisms enable the child to develop proper
understanding of the caregiver’s attention.
It is not surprising that the main goal of the recent, more refined theories of
the origins of joint attention is to identify the mechanisms responsible for, and
the early precursors of the development of joint attention. Nor is it surprising
that these theories do not pose the question of whether the impairment in
joint attention is essentially affective or cognitive. Once we start identifying the
underlying mechanisms for development of joint attention it appears that such
development has links to all three areas of development: affective, cognitive, and
sensory development.
In the more recent theories, then, a common strategy for explaining the
origins of joint attention is to identify its precursors and explain how they
emerge in development. For instance, social orienting is usually taken to be
one of the main precursors of joint attention (for a comprehensive review, see
Dawson et al. 2004). Social orienting refers to children’s ability to „spontaneously
orient to naturally occurring social stimuli in their environment” (Dawson et
al. 2004, p. 272). Usually, children who have difficulties with social orienting,
such as children with autism, also have difficulties with joint attention (Mundy
& Neal 2001). Thus, it has been argued that if the child fails to orient to social
stimuli, that is, if she fails to show appropriate attraction to people, particularly
to the sounds and features of their faces, this will necessarily impair her ability to
engage in joint attentional scenes (Dawson, Meltzoff, Osterling, Rinaldi & Brown
1998; Mundy & Neal, 2001).
The next question is what underlies the child’s ability to orient to social
stimuli successfully. Here, several theories have been proposed. In some instances,
the social orienting is treated as a special case of a more general attentional
functioning (e.g. Bryson, Wainwright-Sharp, & Smith 1990; Courchesne et al.
1994; Dawson & Lewy 1989a, 1989b). There are disagreements about the exact
nature of this special case. According to Courchesne, Chisum, and Townsend
(1995), early social exchanges are quite demanding and require rapid attentional
shifts between different stimuli. In their view, this is exactly what children need
to be able to do if they are to orient spontaneously to social stimuli. Others argue
that the ability to shift attention is not of crucial importance; rather, the very
nature of social stimuli is very complex and requires parallel processing (i.e.
the processing of facial expression, speech, gestures, and the like at the same
time). All of the above seems to be impaired in children with autism. One way to
explain normal processing abilities is by turning to the proper connectivity of the
brain regions. If such connectivity fails to develop the result might be impaired
social orienting, joint attention, language acquisition or in a nutshell the result
could be the onset of autistic spectrum disorder (Just, Cherkassky, Keller, &
Minshew 2004). Furthermore, there are authors who argue that children need to
have a motivational neurological mechanism that assigns a reward value to social
contact if they are to be drawn to social stimuli at all (Mundy 1995; Panskepp
1987). They also argue that this is exactly what children with autism do not have.
It is important to notice that these theories are not mutually exclusive. It
might turn out that the connectivity of brain regions which enables normal
processing as well as a reward mechanism are in play and jointly contribute to
the normal development of social orienting and the processing of emotional
expressions, which together lead to joint attention in children.
Recent studies have shown that typically-developing children first come to
recognize emotional expressions by relying on intersensory perceptions. Later, they
are able to focus on voice or facial expression and to extract emotional meaning
from them alone (Walker-Andrews 1997). What this means is that the ability to
recognize certain expressions depends on the proper integration of stimuli coming
through vision, sound, and touch. Consequently, if the sensory channels do not
work properly, the child will not be able to recognize emotional expressions,
which will thus compromise her ability to grasp the emotional meanings of these
expressions. This also might compromise the development of the neuro-affective
motivation system (reward system) which, in turn, will compromise the child’s
interest in engaging in social interaction, thereby further derailing her social
development. Of course, the causal chain can go in the opposite direction: it
might turn out that the neuro-affective motivation system in the brain allows the
child to enter social interactions, leading to normal brain development and the
connectivity of brain regions. It is important to determine what exactly this causal
chain looks like, and further research in this area would be invaluable.
However, even the empirical findings collected so far point to a very interesting
fact, namely that the child begins to understand emotional expressions and what
they mean far before the child learns language. So, here is the first conclusion we
can make that might contribute to better understanding of the Wittgensteinian
attitudinal approach. It seems that in the first year of a child’s life we can find
the foundation for what is going to become the conceptual link between inner
mental life and outer behavior that some Wittgensteinians have been looking for.
More specifically, empirical findings have shown that if the child is to engage in
joint attentional scenes the child needs to be able to decipher what it means when
mom smiles or when she frowns. Furthermore, if the child is to learn language she
needs to participate in joint attentional scenes. Given that the whole of language
acquisition depends on this preverbal emotional communication we are safe to
assume that the acquisition of mental concepts, the acquisition of the meanings
of such concepts is in fact closely tied to these prelinguistic, nonverbal exchanges.
When the child starts using mental concepts, such as „It hurts’ or ‘I am happy’,
the child starts using them in order to express what she feels. Once language kicks
in the child will not rely on preverbal signals (only) to let others know how she
feels. Moreover, we have seen that in preverbal communication mom’s smile is
her happiness while her frown is disapproval. This is what the child immediately
knows and responds to before she learns language. So, when the child learns
mental concepts this internal connection between smile and joy, or frown and
disapproval is kept. Wittgenstein called this internal connection conceptual. We
can now see that its origins lie in preverbal communication.
Finally, when children develop language they develop the ability to think
about what they think and feel. With language they also develop the ability and
the need to make sense of why we think and feel the way we do. This applies to
their own case as well as the case of others. So, by the time they are four, children
begin to understand what you need to see in order to form a particular belief
and as a result they can pass the false belief test. With language development
they also become aware that there are cases when people lie and start to search
for the indicators that they are (not) lying or pretending in a particular case.
In other words, they start to search for the evidence that their beliefs about
other people’s mental states are correct. But, it is important to notice that this
comes much later when language and abstract capacities are fairly developed.
It is also important to remember that these abilities grow out of immediacy of
preverbal communication where possibility of lying and faking is yet to emerge.
It is interesting to note that Wittgenstein in his Philosophical Investigations
convincingly expressed similar view: „249. Are we perhaps over-hasty in our
assumption that the smile of an unweaned infant is not a pretence?–And on what
experience is our assumption based? (Lying is a language-game that needs to be
learned like any other one.)” (Wittgenstein, Philosophical Investigations, p. 90).
All of the aforementioned insights are of crucial importance for
understanding both the developmental origins of the attitudinal approach as
well as how we become perplexed with the questions what other people feel and
whether they really feel what we think they do. But this worry that others might
lie or pretend does not arise prior to language acquisition nor are we worried
about it in most of our encounters with other people. Even when we start
suspecting that somebody is lying we deal with this particular case separately
from the ordinary cases when we clearly see in other people’s faces or hear in
their words how they feel. The ordinary cases are in a sense a normal background
for thinking about others. As Hyslop correctly notices, to wonder if somebody
is lying or pretending is not the same as questioning that they are persons at all.
However, if we are philosophers it seems to us that we can radicalize and
generalize these exemptions: namely, it seems to us that we can imagine not only
that all people are lying or pretending all the time but also that they don’t have
any feelings and thoughts whatsoever. In other words, it seems to us that we
can conceive of a situation in which other people behave exactly the way we do
but that they have no feelings or thoughts inside at all: i.e. that they are nothing
but machines. Once we radicalize problem in this way we start searching for
some absolute and secure evidence for the existence of other minds. Let me
phrase this in developmental terms. Even though the minds of others seemed
to be fairly unproblematic to a child before she learns language once she learns
language she is equipped to ask all sorts of questions about them including the
very skeptical ones. But, this very possibility tells us more about the nature of
our language than about the nature of other minds. That is, our language is
such that we can make all sorts of distinctions and abstractions the way it suits
us. We can categorize things in many ways and we can also change the rules of
categorization as we go. We can make all sorts of arbitrary distinctions if they are
useful in specific contexts or we can make them just for the fun of it. This is a
precious feature of our language that allows us to investigate nature, be flexible,
adapt well, predict the future, and think about the past. However, it also allows
us to make distinctions such as the one between the inner mental life and the
outer behavior. However, when we take a look at a child’s development the first
thing we notice is that there is no trace of such distinction in our developmental
trajectory. On the contrary, everything that the child does before she learns
language and long after she learns language goes against the assumption that
such distinction makes sense at all.
However, even if this developmental story, that I’ve just told, is sufficiently
convincing to make us embrace the internal, conceptual link between our inner
feelings and behavioral expressions of these feelings there is still one more
question we need to tackle. That is, we still need to see if the Wittgensteinian
attitudinal approach backed with the developmental psychology is sufficient to
dissolve skeptical worries. Let me, now, turn to this question.
3. So far so good but can we deal with the skeptic?

In the previous section I have argued for the psychological primacy of the
Wittgenstenian attitudinal approach toward other minds. This primacy emerges
in the first year of life and enables children to feel with and communicate well
with their caregivers. Through these activities children develop understanding
of others as intentional beings. This is very important finding as it seems that
children even before they learn language, always treat their caregivers and peers
as bodies with minds never as mere bodies. Furthermore, in the previous section I
have also tried to outline how language and concept acquisition emerge from this
attitude toward others. However, we can now ask if this psychological primacy is
sufficient for the successful fight against philosophical skeptic. Hyslop thinks that
it is not. So, let’s go over his worries first and see whether he has a point.
The first question to ask is where the certainty in the existence of other
people’s minds (within the attitudinal approach) comes from. Hyslop argues
that our certainty comes from the fact that we already think of other people’s
behavior as expressing sensation. „We view the behaviour, regard it, apprehend
it, understand it, conceptualize it, see it, as expressive of sensation. That is,
supposedly, why we are certain.” (Hyslop 1995, p.128) As Hyslop goes on to argue
this is probably our normal psychological response to other people. However,
the problem arises when we start invoking regular psychological processes,
mechanisms, and attitudes in order to justify our certainty. Indeed, identified
psychological processes do explain why we feel certain that other people have
minds but they do not justify such certainty. In other words, no matter how
certain we might be about something this still does not mean that that something
is really the case.
Hyslop uses several examples to illustrate how our psychological certainty
can lead us astray. The cases of sexism and racism first come to mind. We all know
that in the patriarchal society women are considered to be inferior. However, this
attitude toward women does not remain on the level of beliefs and thoughts. It
runs deeper. Women are, in fact, perceived as inferior. The inferiority appears on
the observational level. What this means is that when a man sees a woman there
is no need for him to infer that she is inferior. He perceives her like that. A man
does not need to go and look for the inductive nor deductive evidence (he could
if he was challenged to do so). Sexism is very often entrenched in the way people
see the world not just the way they think about the world. The same applies to
racism.
What is striking is that sexists or racists do seem to be certain about their
conception of women or other races and they seem to have the attitude toward
women or other races that is deeper than a theory or a system of beliefs. „For
them the conception is simply of reality. It is completely on a par with seeing a
painting as a painting. It is immediate. It is there, right in front of them, perceived
as such.” (Hyslop 1995, p. 129). Hyslop goes on to illustrate the similarities
between the attitudinal approach to other minds and these other conceptions
that people might develop:
It is not that those with the conception that, say, women’s bodies are for the
delectation of men, are other than misleadingly said to believe this. If they are
thought to believe this the belief will be thought to arise from the conception. The
conception is prior to, and deeper than belief. There is no gap in the conception,
between outer and inner. A woman’s body is seen as for the delectation of men,
that is how they are perceived. That is the immediate experience, as immediate as
in the case of the painting. (Hyslop 1995, p. 129)
To summarize, the aforementioned cases of sexists and racists show that

people can have a particular attitude toward others and that they could develop
particular conceptions about them that are immediate and deeper than mere
beliefs. Such conceptions appear as if they do not leave the space for error.
However, they do. All of the above cases show that even very deep conceptions
might be entirely wrong with no connection to reality. Thus, we can conclude that
these conceptions do provide the feeling of certainty, but they do not justify such
certainty. Hyslop argues that skeptic might say that similar conclusions apply to
our attitude toward other people’s minds. Even though we do have conceptions
about their minds that are deeper than beliefs and that we do indeed perceive
their minds in their behavior skeptic could argue that all of our conceptions
about other people might be entirely wrong. After all: „That women are inferior
is questionable, and questioned, and false. That blacks are inferior is questionable,
and questioned, and false”. (Hyslop 1995, p. 130) Why not assume that something
similar could be the case when it comes to our conception of other minds? Now,
let us examine if the analogy holds the way Hyslop seems to think.
On the structural level it seems that the similarity between the case of other
minds (and our trust in other minds), the case of sexism, and the case of racism
obtains. In all cases we perceive the mind (in the former case) or the inferiority
(in the latter cases) on the very observational level. The very conception of other
people involves minds. For sexists and racists the very conception of women
and/or other races already involves inferiority. But, to look at these cases only on
the structural level is not the only way we can compare them. I will argue that
when we look at the developmental roots of our conception of other people (as
having minds) we can see that the analogy between the case of other minds and
the cases of sexism and racism breaks downs. Now, this is not to say that we will
be safe from skeptical worries altogether. But, at least the skeptical attack will
be postponed a bit and from this new developmental perspective will look less
convincing.
The first thing that Hyslop has not taken into account is that our attitude
toward other people as individuals with minds has developmental priority over
any sort of sexism or racism. If a child and later the adult have trouble ‘reading’
other people minds it is hard to see how the child (or the adult) could be taught
to see others as inferior. To be superior or inferior are social categories that are
built upon the tacit assumption that others exist and have minds, i.e. that they
have some capacity to think and feel.
But there is something more about the primacy of this attitude toward
others that differentiates it from any other attitudes toward others (such as
sexism and racism) that we might develop later. Some empirical findings coming
from the research on autism indicate that the emergence of social cognition
marks the emergence of mind, i.e. the emergence of our inner mental life and
self consciousness.
As we have seen children with autism have problems in many areas of social
cognition. They have problems with social orienting and joint attentional scenes.
Their language acquisition is lagging behind and even when they do learn language
their language remains rigid, their concepts and categories are inflexible (see e.g.
Plaisted 2000, 2001) and their ability to read other people’s minds remains poor.
However, less known but very interesting finding is that inner life along with self
consciousness in these children seems to be impoverished too. Or, as Currathers
(1996) put it, children and individuals with autism do not only have impaired
ability to ascribe beliefs and desires to others but also lack the ability to ascribe
beliefs and desires to themselves. This means that their mind-blindness goes
hand in hand with their lack of self knowledge. Baron-Cohen (1989) found that
autistic subjects find it difficult to remember their own recent false beliefs. They
also seem to have trouble drawing the appearance-reality distinction. If they
don’t have the concept of belief and desire, i.e. if they do not understand that it
is one thing believing that something is the case and yet another that something
is the case they would certainly have difficulty understanding the difference
between their experience and what that experience is of (the reality). Similarly,
Hurlbert et al (1994) found that individuals with Asperger syndrome find it hard
to report their inner experiences. Even those who could pass advanced false
belief tests did not report any inner verbalisation, unsymbolised thinking and
emotional feelings. They only reported visual images. The ones who could not
pass advanced false belief tests did not report any inner experiences at all. These
are indeed striking findings as they indicate that impaired social cognition seems
to be related to impaired self-awareness and impoverished inner mental life.
This means that social cognition and inner mental life are both being constituted
simultaneously through social interaction and preverbal communication during
the first year of child’s life.
The aforementioned cases of children with autism yet again illustrate the
psychological primacy of our attitude toward others over any other conceptions
about others that we might adopt later. It is in fact so important for normal
development that children without it lack a diverse mental life and have
substantially impaired self knowledge. Now, if the child is normally developing it
will remain open and will depend on her immediate surroundings if she is going
to become sexist or racist or liberal minded. But in order to become any of that
she will need to develop the right attitude toward others, the one that will allow
her to treat them as intentional beings with their own minds first.
There is another point that needs to be made here. The findings that indicate
that our own mind is being constituted along with our knowledge of other
people’s mind is one of the single most important insights of developmental
psychology. The sceptical worry about other people’s minds starts off with
the assumption that our own mind along with our self-knowledge comes
first while our knowledge of others is somehow derivative and comes later.
However, developmental psychology is telling us that developmentally this is
not the case. The studies on autism have shown that there is no mind of ours
or self-knowledge without knowledge of other minds and that they are being
constituted simultaneously in the first year of child’s life. Thus, it seems that
there is no impairment in social cognition without the profound impairment
in self awareness and affective development. The question now is whether these
findings and the above points could satisfy our philosophical sceptic.
I have shown that we could answer Hyslop’s worries that the attitudinal
approach is similar to sexism or racism in a sense that it gives us the feeling
of certainty but that it may as well be unfounded as the other two. I have
pointed out that there is developmental disanalogy (if not structural disanalogy)
between attitudinal approach to other people’s minds, on one hand, and sexism
and racism, on the other. This means that if they are at least developmentally
different and if the attitudinal approach has the primacy over other two we don’t
have to worry about that it might turn out that we are wrong about other people’s
minds in the same way as many have been shown to be wrong about inferiority
of women or other races. Furthermore, I have argued that in developmental
terms there is no asymmetry between our own mind and self-knowledge on one
hand and other people’s minds and our knowledge of their minds, on the other.
As developmental science has shown both kinds of knowledge emerge in the
same time. But, is this developmental dispersal of asymmetry sufficient to fight
the sceptic? The answer must be no. However, some conditions apply. So, let me
conclude with some thoughts about the nature of the sceptical endeavour.
As we have seen in the previous section our language is such that it enables
us to abstract from many situations as well as to imagine and conceive of many
novel ones. So, to all of the above findings from psychology the skeptic might say
that they are of no help because we can still imagine situation in which all people
around us are just robots. We can imagine that such robots have raised us and
that we have passed though all of the necessary stages of normal psychological
development which includes the development of our proper attitude toward
others as fellow human beings and persons. But, in this imagined situation this
attitude would be misleading as these others would have no inner mental life
whatsoever. All of this we can imagine so the skeptic still has the space to remain
skeptical should she decide to do so. However, as I said some conditions apply.
First, it is not clear if the imagined situation is consistent at all. A thought
experiment of this kind could go astray. While philosophers are inclined to argue
that they can imagine all sorts of things it might turn out that imagining is more
demanding than they have initially thought. The famous ‘brain in a vat’ thought
experiment has been challenged in this way (Thompson and Cosmelli, 2011).
Thompson and Cosmelli have argued that once we start imagining the very
details of how the brain in a vat functions it turns out that such brain would not
be able to function without something like an operational body. So, for instance,
in order to be alive such brain would need to have some sort of metabolism (the
exchange of nutritive elements and oxygen). Also, simulating perception would
require something like an operational eye (see e.g. A. Clark 2001) and so on. All
of this would require some sort of body. However, to imagine a brain without the
body was the sole purpose of this thought experiment. In the case of the thought
experiment of our skeptic it remains to be seen if developmental psychology
could play a role similar to the role that neurophysiology played in ‘the brain
in a vat’ scenario. But, even if we cannot make such a strong case against our
skeptic (as Thompson and Cosmelli made against theirs) we should still hold
on to the insights about the development of human mind, social cognition, and
language acquisition that we have gained from developmental psychology. These

insights tell us that developmentally speaking there is no asymmetry between
our self knowledge and knowledge of others and that our language emerges
from preverbal communication. Now, this piling empirical evidence does tell us
something about the inherently social nature of human minds and the origins of
the meanings of mental concepts. Once we learn this lesson it is harder to take
a sceptical attack seriously. Unless the skeptic gives us some additional reasons,
not just imaginable situation, for doubting the existence of other minds we are
safe to keep believing that they exist.
References
Avramides, A., (2001) Other Minds, London: Routledge
Baron-Cohen, S. (1989). „Are autistic children behaviorist? An examination of
their mental-physical and appearance-reality distinctionis”. Journal of autism
and developmental disorders, 19, 579–600.
Baron-Cohen, S. (1995). Mindblidness: An essay on autism and theory of mind.
Cambridge: Bradford/MIT Press.
Baron-Cohen, S., Leslie, A.M., & Frith, U. (1985). „Does the autistic child have a
„theory of mind”?”, Cognition, 4, 37–46.
Bryson, S.E., Wainwright-Sharp, J. A., & Smith, I.M. (1990). „Autism: A
developmental spatial neglect syndrome?”. In J Enns (Ed.), The development
of attention: Research and Theory (pp. 405–427). North Holland: Elsevier.
Carruthers, P., (1996). ”Autism as Mind-Blindness: an elaboration and partial
defence”. P. Carruthers & (Eds.), P.K.Smith, Theories of theories of mind (pp.
257–273), Cambridge University Press.
Charman, T. (2003). „Why is joint attention a pivotal skill in autism?”. Philsophical
Transactions of the Royal Society of London B, 358, 315–324.
Clark, A. (2001). Mindware: An Introduction to the Philosophy of Cognitive
Science. New York: Oxford University Press.
Courchesne, E., Townsend, J., & Akshoomoff, N. A. (1994). „Impairment in
shifting attention in autistic and cerebellar patients”. Behavioral Neuroscience,
108, 848–865.
Dawson, G., Lewy, A. (1989a). „Arousal, attention, and the socioemotional
impairments of individuals with aurism”. In G. Dawson (Ed.), Autism: Nature,
diagnosis, and treatment (pp. 49–74). New York: Guilford Press.
Dawson, G., Lewy, A. (1989b). „Reciprocal subcortical-cortical influences in
autism: The role of attentional mechanisms”. In G. Dawson (Ed.), Autism:
Nature, diagnosis, and treatment (pp. 144–173). New York: Guilford Press.
Dawson, G., Meltzoff, A., Osterling, J., & Rinaldi, J. (1998). „Neurophychological
correlates of early symptoms of autism”. Child Development, 69, 1276–1285.
Dawson, G., Toth, K., Abbott, R., Osterling, J., Munson, J., Estes, A. & Liaw,
J. (2004). „Early social attention impairments in autism: Social orienting,
joint attention and attention to distress”. Developmental Psychology, 40, 271–283.
Ellis, A.E., & Oakes, L.M. (2006). „Infants flexibly use different dimensions to
categorize objects”. Developmental Psychology, 42, 1000–1011.
Hampshire, S., (1970). „The analogy of feeling”. in T.O. Buford (ed.), Essays on
other minds (pp. 9–23). University of Illinois Press.
Hacker, P.M.S., (1972). Insight and illusion: Wittgenstein on philosophy and the
metaphysics of experience, Clarendon Press
Hobson, P. (1993). Autism and the development of mind. Hove, UK: Lawrence
Erlbaum Associates, Ltd.
Hyslop, A., (1995). Other Minds, Dordrecht/Boston/London, Kluwer Academic
Publishers
Hurlbert, R.T., Happe, F., & Frith, U.,(1994). „Sampling the form of inner
experience in three adults with Asperger syndrome”. Psychological Medicine,
24, 385–395.
Just, M. A., Cherkassky, V. L., Keller, T. A., & Minshew, N. J. (2004). „Cortical
activation and synchronization during sentence comprehension in high-
functioning autism: Evidence of underconnectivity”. Brain, 127, 1811–1821.
Johnson, S. C. (2000). „The recognition of mentalistic agents in infancy”. Trends
in Cognitive Sciences, 4, 22–28.
Kasari, C., Sigman, M., Mundy, P. & Yirmiya, N. (1990). „Affective sharing in the
context of joint attention interaction of normal, autistic and mentally retarded
children”. Journal of Autism and Developmental Disorders, 20, 87–100.
Kojen, L., (2009). Ogled iz filozofske psihologije, Čigoja Štampa
Kripke, S.A., (1982). Wittgenstein on Rules and Private Language, Blackwell
Leekam, S. R., & Moore, C. (2001). „The development of attention and joint
attention in children with autism”. In J.A. Burack, T. Charman, N. Yirmiya,
& P.R. Zelazo (Eds.) The Development of autism: Perspectives from theory and
research (pp. 105–31). New Jersey: Lawrence Erlbaum Associates.
Loveland, K., & Landry, S. (1986). „Joint attention and language in autism
and developmental language delay”. Journal of Autism and Developmental
Disorders, 16, 335–49.
Malcolm, N., (1958). „Knowledge of Other Minds”, Journal of Philosophy, vo1.55,
969–78.
Malcolm, N., (1959). Dreaming, London: Routledge
Markman, E., (1989). Categorization and naming in children. Cambridge, Mass.:
The MIT Press.
Mill, J.S., (1872). An Examination of Sir William Hamilton’s philosophy, (fourth
edition), London: Longman, Green, Reader, and Dyer
Miller, Alexander and Crispin Wright, eds., (2002). Rule-Following and Meaning.
Montreal and Kingston: McGill-Queens’s University Press
Mundy, P. (1995). „Joint attention and socio-emotional approach behavior in
children with autism”. Development and Psychopathology, 7, 63–82.
Mundy, O., Sigman, M., Ungerer, J., & Sherman, T. (1986). „Defining the social
deficits in autism: The contribution of non-verbal communication measures”.
Journal of Child Psychology and Psychiatry and Allied Disciplines, 27, 657–669.
Mundy, P. & Crowson, M. (1997). „Joint attention and early social

communication: Implications for research on interventions with autism”.
Journal of Autism and Developmental Disorders, 6, 653–676.
Mundy, O., & Neal, R. (2001). „Neural plasticity, joint attention and a transactional
social-orienting model of autism”. In L. Glidden (Ed.), International review
of research in mental retardation: Vol. 23. Autism (pp. 139–168). New York:
Academic Press.
Panskepp, J. (1986). „The neurochemistry of behavior”. Annual Review of
Psychology, 37, 77–107.
Paparella, T., D’Angiola., & Kasari, C. (2001). „The emergence of joint attention
skills, eye-contact, and positive affect in young children with autism”.
Poster presented at the 2001 meeting of the Society for Research in Child
Development, Minneapolis, MN.
Paparella, T., & Kasari, C. (2002). „The communicative quality of joint attention
and requesting skills in young children with autism”. Manuscript submitted
for publication.
Plaisted, K. (2000). „Aspects of autism that theory of mind cannot explain”. In
S. Baron-Cohen, H. Tager-Flusberg, & D. Cohen (Eds.) Understanding other
minds (pp. 224–250). New York: Oxford University Press.
Plaisted, K. (2001). „Reduced generalization in autism: An alternative to weak
central coherence”. In J.A. Burack, T. Charman, N. Yirmiya, & P.R. Zelazo
(Eds.) The Development of autism: Perspectives from theory and research (pp.
149–171). New Jersey: Lawrence Erlbaum Associates.
Russell, B., (1970). „Analogy”, in T.O. Buford (ed.), Essays on other minds (pp.
3–9). University of Illinois Press.
Shanker, S., (2013). Calm, alert, and learning, Pearson Canada Inc.
Tantam, D., (2009). Can the world afford autistic spectrum disorder? Nonverbal
communication, asperger syndrome, and the interbrain, London: Jessica
Kingsley Publishers.
Thompson, E., Cosmelli, D., (2011).”Brain in a vat or body in a world? Brainbound
versus enactive views of experience”, Philosophical topics, 39, 163–180.
Ter Hark, M.R.M., (1991), „The Development of Wittgenstein’s Views about the
Other Minds Problem”, Synthese, 87, 227–53.
Walker-Andrews, A.S. (1997). „Infants’ perception of expressive behaviors:
Differentiation of multimodal information”. Psychological Bulletin, 121, 437–456.
Wimmer, H. J. Perner, J. (1983). „Beliefs about beliefs: Representation and
constraining function of wrong beliefs in young children’s understanding of
deception”. Cognition 13 (1), 103–128.
Wittgenstein, L., (1953), Philosophical Investigations, Basil Blackwell.
Drago Đurić Original Scientific Paper
University of Belgrade UKD 141.4:165.15
28–21:165.15
141.4 Ал-Кинди
141.4 Крејг В. Л.
ALKINDI’S AND W. L. CRAIG’S

COSMOLOGICAL ARGUMENTS1
Abstract: In this paper I shall consider the similarities and differences in the
cosmological argumentation on the existence of God between the 9th-century
Muslim philosopher al-Kindi and contemporary Christian philosopher William
Lane Craig. My focus here will not be on the value and soundness of their
argumentation, but only on the structure and type of their arguments. The credit
for this argumentation’s reappearance in the modern thought belongs to Craig, who
referred to it, above all, in his book Kalam Cosmological Argument (1979). The
basic elements of their general form of argument can be found in the works of the
6th-century Christian philosopher John Philoponus. All major steps of Craig’s kalam
cosmological argumentation can be found in al-Kindi’s works. Bearing in mind his
large opus on the kalam cosmological argument, it is surprising to see Craig writing
that this argument is „extremely simple”, and has the form as follows: 1. Everything
that begins to exist has its cause. 2. The world began to exist. 3. Therefore, the world
has its cause. Generally considered we might think that a better way to express this
argumentation would follow three major steps: 1. First, it has to be proved that
the world is not eternal, or that it began to exist. 2. Second, that the world does
not come to exist by itself, but that it has a cause of its beginning. 3. Third, that
cause of the beginning of its existence is an orthodoxly conceived monotheistic God.
Philoponus gives his arguments only for the first two steps of this argumentation.
Al-Kindi presents his arguments for all three steps. Craig, in essence, repeates al-
Kindi’s arguments, but adds some convergent arguments that are based on two
contemporary scientific theories, unknown during Philoponus and al-Kindi’s times:
Big Bang cosmological theory and the second law of thermodyamics.
Here we shall consider some similarities and some differences in the

cosmological argumentation for the existence of God between 9th-century
Muslim philosopher al-Kindi and contemporary Christian philosopher William
1 This paper was presented at 8th Annual International Conference on Philosophy 27–30
May 2013, Athens, Greece. The paper was created within the projects “Dynamical systems
in nature and society: the philosophical and the empirical aspects” (179041) and “Logical
and epistemological basis of science and metaphysics” (179067) supported by Ministry of
Education, Science and Technological Development of Republic of Serbia.
Lane Craig. Our focus here will not be on the value and soundness of their
argumentation, but only on the structure and type of their arguments. The
medieval kalam cosmological argument, which in its complete structure begins
with al-Kindi’s works, has almost been forgotten in the modern West. All credit
for its new life belongs to Craig, who referred to it, above all, in his book Kalam
Cosmological Argument (1979).2
Bearing in mind his large opus about the kalam cosmological argument, it is
surprising that Craig writes that this argument is „extremely simple”, and has the
fallowing general form:
1. Everything that begins to exist has a cause.
2. The world began to exist.
3. Therefore, the world has a cause.
It is probable that this form of argument was explicitly presented for the
first time by al-Ghazali, who, in his book Kitab al-Iqtisad, writes: „Every being
which begins has a cause for its beginning; now the world is a being which
begins; therefore, it possesses a cause for its beginning” (Al-Ghazali, 1963: 15–
16). Rudimentary elements of this general form of argument we can find in the
works of 6th-century Christian thinker John Philoponus.
But the general form of argument is not simple at all, and is definitely
not „extremely simple”. This argument has a complex structure containing
many subarguments. We shall now present the general structure of the kalam
cosmological argumentation, which one can recognize in both al-Kindi, and
Craig. I think that, rather, this argument develops trough three major steps:
1. First, it must be proved that the world is not eternal, or that it began to
exist.
2. Second, that world does not come to exist by itself, but it has a cause of
its beginning.
3. Third, that the cause of the beginning of its existence is monotheistic God.
Historical significance of al-Kindi’s argumentation

Before we move to a comparison, we can first say something about the
historical significance of al-Kindi’s argumentation. Aristotle, especially in the
Physics VIII, and On the Heavens I, gives developed argumentation for and against
the eternity of the world. The result of his argumentation was the conclusion that
world is eternal. Christian philosopher John Philoponus made the first attack on
Aristotle’s solution. His criticism of Aristotle was directed against the coherence
between two of his theses; that actual infinity is impossible, and that the world a
parte ante is eternal.
Philoponus gives some significant contributions to the first step of
argumentation with which we are come to the conclusion that world is not
2 Craig, L. W., (1979), Kalam Cosmological Argument, London: Macmillan & Co.
Drago Đurić: Al-Kindi’s and W. L. Craig’s cosmological arguments 159
eternal or that it began to exist. Some of his arguments are repeated in writings
of medieval Muslim and Jewish thinkers (Davidson, H., 1969). It seems that he
believes that it is self-evident that the world is not cause of itself, and that the
cause of the beginning of the world is a monotheistic conceived God.
Al-Kindi was, it seems first thinker who gave the full structure of the kalam
cosmological argument. He had developed his own argument for the finitude of
world, an argument for the thesis that the world has a transcendent cause (or
that world is not cause of itself), and argument that the cause of the beginning of
the world is the God of monotheism. These are essentially the same steps, which
make up the parts of Craig’s contemporary kalam cosmological argumentation.
Because of this, their arguments are suitable for comparison.
Characteristics of al-Kindi arguments

Although Philoponus in some way bases his arguments on mathematics as
well, in all of his arguments he sticks to the thesis that actual infinity is impossible,
and explicitly uses examples which have no mathematical character. Except in
his most famous work, On the First Philosophy, al-Kindi also gives mathematical
proofs for the impossibility of actual infinity in some of his shorter epistles: On
What cannot be Infinite and of What Infinity May be Attributed (Al-Kindi, 1975),
On Divine Unity and the Finitude of the World’s Body (Al-Kindi, 2007) and On
the Finitude of the Universe (Al-Kindi, 1965).
There are many points where al-Kindi speaks about the mathematical
character of the first step of his argumentation. First, the objects of his
argumentation are homogeneous magnitudes, second, his argumentation begins
with axioms or with, as he says, „premises that are first, evident, and immediately
intelligible”, and, lastly, his favored type of argumentation is argumentum reductio
ad absurdum.
In an essay On the Finitude of the Universe, the mathematical characteristic of
entities with which he operates in argumentation is explicated more clearly than
in other works. It is in general implicitely assumed. In the forementioned essay he
explicitly says that the object of argumentation will be „magnitudes”. Let us see
how he defines the concept „magnitude”. He writes: „When we say ‘magnitude’
in this discipline (mathematics) we mean by it only one of three things. Either
(1) that which has length only, I mean by this the line, or (2) that which has
length and breadth only, I mean by this the surface (plane), or (3) that which
has length and breadth and depth, I mean by this the body”. (Al-Kindi, 1965:
187) Here we can operate, he claims, only with „homogeneous magnitudes”,3
and under homogeneous magnitudes he means those „magnitudes which are
altogether lines or altogether surfaces or altogether bodies”. Nevertheless, it
seems that al-Kindi has in mind impossibility of mathematical operations on
non-homogeneous magnitudes. We, for example, can not collect lines and
bodies, or subtract a body from a surface.
3 One can see the concept of “homogeneous magnitude” in Euclid’s Elements (Book. 5. Def. 3.).
In his works On first philosophy and On Divine Unity and the Finitude of the
World’s Body, the homogeneous magnitude under consideration is the body. In
the work On the Finitude of the Universe the object of argumentation is a pure
concept of homogeneous magnitude. In On What cannot be Infinite and of What
Infinity May be Attributed the objects under consideration are just „something”
or „thing”. In accordance with all that he says, it seems that al-Kindi is motivated
by didactic reasons. Namely, in the latter essay al-Kindi writes: „I am penning
here what is in my opinion sufficient on view of the extent of your knowledge.”
(Al-Kindi, 1975: 194)
First step of argumentation: finitude of the world

The proposition „the world began to exist” implies that the world is not
eternal or that it has no infinite past. If we prove that the world does not have
an infinite past, we have proved that it began to exist. Let us first see how Craig
formulates this argument. He usual provides us with two kinds of arguments.
The first set of arguments he calls metaphysical, or philosophical, or deductive,
and they are: 1) the argument based on „the impossibility of the existence of an
actual infinite”; and 2) the argument based on „the impossibility of forming an
actual infinite by successive addition”. The second set of arguments Craig calls
the inductive or scientific arguments, and they are: 1) the argument based on
the Big Bang cosmological theory, and 2) the argument based on the second
law of thermodynamics. We will consider only metaphysical or philosophical
arguments, because the Big Bang cosmological theory and second law of
thermodynamics were unknown during al-Kindi’s time. At first glance, it seems
that the two argument of the first set are similar, but both al-Kindi and Craig
claim that they are independent.
The first argument of the first set of argumentation, based on the impossibility
of the existence of an actual infinite, has the following general form:
1. An actual infinite cannot exist.
2. An infinite temporal regress of events is an actual infinite.
3. Therefore, an infinite temporal regress of events cannot exist.
The soundness of the whole argument depends on the first premise. For the
second premise, we need only to recognize that the infinite temporal regress of
events is an actual infinite.
Craig usually only defends the impossibility of existence of something infinite
in the real spacio-temporal world from the theory of the infinite sets, which was
developed from 19th-century mathematician Georg Cantor’s consideration of
this issue. He also gives examples in which he tries to show that any attempt
to realize infinity in the spacio-temporal world leads to contradiction or to the
absurd. For that purpose, Craig often refers to David Hilbert’s view, given in
his paper On the Infinite, and quotes following, well-known words: „The infinite
is nowhere to be found in reality. It neither exists in nature, nor provides a
legitimate basis for rational thought” (Hilbert, D. 1964: 151).
One of the most interesting parts of al-Kinidi’s opus is his attempts to give a
proof to the first premise of the above-mentioned argument. As we have said, his
argumentation begins with axioms. Despite al-Kindi presenting these axioms in
different variations, we shall cite one presented in On the First Philosophy. This
„true first premises which are thought with no mediation” are, he writes, the
following:
1. All bodies of which one is not greater than the other are equal.
2. Equal bodies are those where the dimensions between their limits are
equal in actuality and potentiality.
3. That which is finite is not infinite.
4. [When] a body is added to one of equal bodies it becomes the greatest
of them, and greater than it had been before that body was added to it.
5. Whenever two bodies of finite magnitude are joined, the body which
comes to be from both of them is of finite magnitude.
6. The smaller of every generally related thing is inferior to the larger, or
inferior to a portion of it (Al-Kindi, 1974: 114).
After presenting these axioms al-Kindi goes on to the argument itself. The
type of argument that he applies to the object of argumentation – the body –
is reductio ad absurdum argument. Al-Kindi, as is usual in the case of reductio
ad absurdum, begins with an assumption of a reductio, which is the following:
„Now if there is an infinite body, then whenever a body of finite magnitude is
separated from it, that which remains of it will either be a finite magnitude or an
infinite magnitude.” In this type of argument, it is usually sufficient to prove that
one of the disjuncts is contradictory or that it leads to a contradiction. In that
case, the other disjunct will necessarily be true. Al-Kindi tries to demonstrate
that both disjucts are contradictory or absurd.
The argument for the first disjunct is presented in this way: „If that which
remains of it is a finite magnitude, then whenever that finite magnitude which is
separated from it is added to it, the body which comes from them both together
is a finite magnitude [ax. 5]; though that what comes to be from them both is
that, which was infinite before something was separated from it. It is thus finite
and infinite, and this is an impossible contradiction” (Al-Kindi, 1974: 115).
The second disjunct is divided into a new disjunction, and al-Kindi’s reductio
here takes the following form: „If the remainder is an infinite magnitude, then
whenever that which was taken from it is added to it, it will either be greater
than or equal to what it was before addition”. The argument for the absurdity of
the first subdisjunct of the second disjunct goes as follows:
If it is greater than it was, than that which has infinity, the smaller of two
things being inferior to the greater, or inferior to the portion of it – if the smaller
body is inferior to the greater, then it most certainly is inferior to a portion of it
– and thus the smaller of the two is equal to a portion of greater. Now, two equal
things are those whose similarity is that the dimensions between their limits are
the same, and therefore the two things possess limits – for ‘equal’ bodies which
are not similar are those (in) which one part is numbered the same, though (as a
whole) they differ in abundance or quality or both, they (too) being finite – and
thus the smaller infinite object is finite, and this is an impossible contradiction,
and one of them is not greater than the other. (Al-Kindi, 1974: 115)
The second subdisjunct of the second main disjunct al-Kindi presents in this way:
If it is not greater than that which was before it was added to, a body having
been added to a body and not having increased anything, and the whole of this
is equal to it alone–it alone being a part of it–and to its (own) part, which two
(parts) join, then the part is like the all, (and) this is impossible contradiction.
(Al-Kindi, 1974: 116)
The aim of this argumentation is to give proof that the world a parte ante is
not infinite. Al-Kindi thinks that by this argumentation he „has been explained
that any quantitative thing cannot have infinity in actuality”, and, since time
is quantitative, it cannot have infinity in actuality. Therefore, time must have
a finite beginning. Al-Kindi comes to the conclusion that the world a parte
ante is finite through the principle of coextensivness, according to which time,
motion and body are coextensive. We should put aside al-Kindi`s sophisticated
argumentation regarding the coexstensivness of time, motion, and body. It is
sufficient here to say that al-Kindi comes to the conclusion that the world a
parte ante is finite, through the coextensivness of time and the ‘body of whole’
(universe). Because time is duration of the body, there cannot exist any body
without time. Thus, if time is a parte ante finite, than the body of whole or the
universe is a parte ante finite also. In this way, we come to the conclusion that
the world begins to exist (Al-Kindi, 1974: 116–121; 2007: 143–145).
Let us now see how Craig comes to this same conclusion. After proving that
the actual infinite is only a conceptual idea, and that it cannot be something
that exists in reality, Craig claims that an infinite temporal regress of events is
an actual infinite. Therefore, an infinite temporal regress of events cannot exist.
But because the temporal existence of world a parte ante is nothing other than
the temporal series of past events of the world’s existence, than the world must
begin to exist.
The second argument of the first set of argumentation, the argument based
on the impossibility of forming an actual infinite by successive addition, Craig (with
Sinclair) calls metaphysical, or philosophical, or deductive arguments (Craig, L.
W. and Sinclair, D. J., 2009: 103). This argument has the following general form:
1. The temporal series of events is a collection formed by successive
addition.
2. A collection formed by successive addition cannot be an actual infinite.
3. Therefore, the temporal series of events cannot be an actual infinite.
At the beginning of the presentation of this argument, in On the First
Philosophy, al-Kindi clearly suggests that it is an independent argument by
saying: ‘Let us now explain it in another way that it is not possible for time to
have infinity in actuality’ (Al-Kindi, 1974: 121). He again considers this issue
reductio ad absurdum, but now in classical form according to which the absurdity
of one disjunct implies that the other disjunct becomes true. Issue by means of
a reductio is the next disjunction: past time is either infinite or finite. Here he
assumes the coextension of the time and body. Al-Kindi defends the thesis that
past time is finite by showing that other disjunct is subject to contradiction or
absurdity. This argument, already known to Philoponus, al-Kindi presents in
the following way: „If a definite time cannot be reached until a time before it is
reached, nor that before it until a time before it is reached, and so to infinity; and
the infinite can neither be traversed nor brought to an end; than the temporally
infinite can never be traversed so as to reach a definite time” (Al-Kindi, 1974:
121–122). In other words, if in its current state world must have traversed a
infinite number of previous states, than, because infinity cannot be traversed, the
current state of the world would never exist. But the current state of the world
does exist. Therefore, the world has a finite number of previous states, and it has
no infinite past. Consequently, there is a beginning of the world’s existence.
Craig also says that this argument is independent because it „denies that the
collection containing an actual infinite number of things can be formed by adding
one member after another.” The core of Craig argumentation is very similar to al-
Kindi’s argument. Let us see how this looks in Craig’s version. He writes:
[B]efore the present event could occur, the event immediately prior to it
would have to occur; and before that event could occur, the event immediately
prior to it would occur; and so on ad infinitum. One gets driven back and
back into the infinite past, making it impossible for any event to occur. Thus,
if the series of past events were beginningless, the present event could not have
occurred, which is absurd. (Craig and Sinclair, D. J., 2009: 118).
Second step of argumentation: was the world’s

beginning caused or is it the cause of itself?
In the second step of the argumentation, one must prove that world is
not the cause of its own beginning, but that this cause transcends it. What is
significant is that presentation of that problem in On the First Philosophy comes
immediately after the previous step of the argumentation. At the beginning of
The Third Chapter of the First Part, al-Kindi says: „An investigation whether it
is or is not possible for a thing to be the cause of the generation of its essence,
shall now follow the previous (discussion)” (Al-Kindi, 1974: 123). Because we
are here presenting the structure of the whole argument we cannot discuss al-
Kindi’s sophisticated dialectics of essence and existence, whose main conclusion
is that no one thing can be the cause of itself. Consequently, the world cannot
be cause of itself, and, therefore, it must have been brought into existence by
something that transcends it.
Al-Kindi, as we will see, shows that the inference from the conclusion
that the world has a transcendent cause to the conclusion that that cause is a
monotheistic God is not immediate. It requires a further step, which, according
to our presentation of the structure of the entire argumentation, is the last step
of the argument.
Before we come to this last step in the structure of the kalam cosmological
argument, let us say something about Craig’s consideration of the second step.
In many of his shorter presentations Craig simply assumes the causal principle
expressed in the first premise of the „simple” general form of the argument. Craig but
usually considers this question in a discussion regarding the Big Bang cosmological
hypothesis (see Craig, L. W. and Smith, Q., 1993). In many of his considerations of
this question Craig rejects the hypothesis that world begins its existence uncaused,
spontaneously, or that it is cause of its own beginning, and he offers us arguments
for the hypothesis that there was a cause of the world’s beginning. His arguments
usually have a probabilistic character. His argument for the hypothesis that there
was a cause of the world’s beginning is more probable than other hypotheses. In one
of his last declarations on this question he says that the principle expressed in the
premise „everything that begins to exist has a cause” is „obviously true – at least,
more plausibly true than its negation” (Craig, 2009: 182).
In his conclusion, Craig gives us three closely connected reasons for this.
First, he defends an old metaphysical principle ex nihilo nihil fit. In other essays,
he writes that, „if absolutely nothing existed prior to the Big Bang – no matter,
no energy, no space, no time, no deity – , than it seems impossible that anything
should begin to exist” (Craig, 1993: 627). It is thus more probable that there
was a cause of the world’s beginning. For the second reason he asks why only
universe? Namely, if the causal principle is valid for all other cases in the world,
then why would it not be valid for the beginning of the world itself? Third,
Craig’s reason rests on the experiential confirmation. The general validity of the
causal principle (validity without exception) has its confirmation, thinks Craig,
in both our everyday life and scientific experiment. With regard to the general
validity of the causal principle in science he cites the „naturalistic philosopher
of science” Bernulf Kanitscheider, who writes that the rule ex nihilo nihil fit is a
„metaphysical hypothesis which has proved so fruitful in every corner of science
that we are surely well-advised to try as hard as we can to eschew processes of
absolute origin” (Kanitscheider, 1990: 344). The validity of the causal principle,
concludes Craig, is more probable then its negation. Bing bang cosmology does
not allow for material cause, because before the big bang no matter, energy,
space, or time exists. However, in accordance with his probabilistic mode of
argumentation, Craig writes: „For if coming into being without a material cause
seems impossible, coming into being with neither a material nor an efficient
cause is doubly absurd” (Craig and Sinclair, 2009: 189).
Third step of argumentation:

the cause of the beginning of the existence of the world
is a monotheistically concieved God
In this step we see argumentation for the thesis that the cause of the
beginning of the world is a monotheistically concieved God. The God of
classical monotheism is usually described by its attributes. The qustion is, at
which attributes can we arrive at through conceptual analysis of the cause of
the beginning of the world? The first attribute of God is that God is one. Al-
Kindi offers us an argument for the oneness of God in On the First Philosophy
and On Divine Unity and the Finitude of the World’s Body. Here we will present
his argument according to his last work, although in this work he uses the word
‘Creator’. It is, in the flow of his entire argumentation, too „early”. Namely, with
little bit of benevolence to al-Kindi`s opus we can find a relatively developed
argument for a oneness of a cause, and than an argument that this cause is a
Creator or, as he says, an agent. Instead of Creator we will here use the word
cause. Applying the thesis of impossibility of infinite regress to this issue, al-
Kindi in his argumentation starts with the disjunction that this cause must be
either one or many. If they are many then they must be composite. Compositions,
writes he, „have a composer”. If it is a many, then it must have a composer, etc. In
this way then there is „something [that is, the sequence of agents] that is actual
infinite – but the falsity of this has been explaind” (Al-Kindi, 2007: 147–148).
Therefore, the first cause must be one.
Craig basically repeats al-Kindi’s argumentation. He writes that the first
cause must be uncaused, „since [...] an infinite regress of causes is impossible”.
Although „one could [...] arbitrarily posit a plurality of causes in some sense
prior to the origin of the universe, but ultimately, if the philosophical kalam
arguments are sound, this causal chain must terminate in a cause which is
absolutely first and uncaused”. Craig adds the principle of Ockham’s razor to al-
Kindi’s argumentation, which, as he thinks, „dictates that we are warranted in
ignoring the possibility of a plurality of uncaused causes in favor of assuming the
unicity of the first cause” (Craig and Sinclair, 2009: 192).
The other property of the first cause to which both al-Kindi and Craig refer
is that this cause is personal. Al-Kindi, in the On the First Philosophy, further
concludes that the first cause is the True One. This True One is „the cause from
which there is a beginning of motion”. That which sets something in motion
is, he says, „the agent” (Al-Kindi, 1974: 162). In The One True and Complete
Agent and the Incomplete Metaphorical „Agent” he writes that „first true action
is bringing being into existence from nonbeing”. This action is properly God’s
action. Only to this action properly applies the term „creation”. Namely, „the true
agent acts upon what is affected without itself being acted in any way” (Al-Kindi,
2007a: 169). The True Agent, as the first, is the real and last origin of motion.
Because of this, it can be called a Creator.
In Craig’s language, the action of al-Kindi’s agent is, basically, at the personal
action. Craig (with Sinclair) offers us three reasons for the personal character of
the first transcendent cause. Let us present them in short. First, he cites Richard
Swinburne (Swinburne, 2004, 35–38) who claims that there are two types of
causal explanation: scientific explanation explicated in the language of laws and
initial conditions and personal explanation in terms of agents and their volitions.
According to the Big Bang hypothesis, the first state of the universe „cannot
have a scientific explanation, since there is nothing before it, and therefore, it
cannot be accounted for in terms of laws operating on initial conditions” (Craig
and Sinclair, 2009: 193). Second, to the personal character of the First Cause,
we arrive, suggests Craig, by a conceptual analysis of it. With that analysis, it
is not hard to show, he writes, that the first cause must be an immaterial,
beginningless, uncaused, timeless, and spaceless being. He finds that there are
only two candidates we can describe with these attributes: either abstract objects
or unembodied minds. Abstract objects (like numbers, sets, propositions, and
properties) cannot be first cause. Therefore, cause of the beginning of the world
can only be an unembodied mind. Third, that the First Cause has a personal
character is also implied, he writes, „by the fact that only personal, free agency
can account for the origin of a first temporal effect from a changeless cause”
(Craig and Sinclair, 2009: 193).
In the end, we can conclude that Al-Kindi’s kalam cosmological argument
has all the elements contained in Craig’s more developed argument. The main
differences are in the additional convergent and confirmatory arguments, which
are developed in contemporary science and philosophy.
References
Al-Kindi (1974). On the First Philosophy (intr. and comm. Ivry, L. A). In: Al-
Kindi’s Metaphysics, Albany: State University of New York Press.
Al-Kindi (1975). On What cannot be Infinite and of What Infinity May be
Attributed, in: Islamic Studies Vol. 14. No. 2. (transl. Shamsi, A. F.)
Al-Kindi (2007). On Divine Unity and the Finitude of the World’s Body, in:
Classical Arabic Philosophy – An Anthology of Sources, (ed. and transl.,
McGinnis, J. & Reisman, C. D.), Hackett Publishing Company, Indianapolis/
Cambridge.
Al-Kindi (2007a). The One True and Complete Agent and the Incomplete
Metaphorical „Agent”, in: Classical Arabic Philosophy – An Anthology of
Sources, (ed. and transl., McGinnis, J. & Reisman, C. D.), Hackett Publishing
Company, Indianapolis/Cambridge.
Al-Kindi (1965). On the Finitude of the Universe, in: Isis, Vol. 56, No. 4. (transl.
Rescher, N. & Khatchadourian, H.)
Craig, L. W., (1979), Kalam Cosmological Argument, London: Macmillan & Co
Craig, W. L. (1993). „The Caused Beginning of the Universe: A Response to
Quentin Smith”, British Journal for the Philosophy of Science 44.
Craig, L. W. and Sinclair, D. J. (2009). „The Kalam Cosmological Argument”, in:
The Blackwell Companion to Natural Theology (ed. Craig, L. W. and Moreland,
P. J.) Oxford, Wiley-Blackwell
Craig, L. W. and Smith, Q. (1993). Theism, Atheism, and Big Bang Cosmology,
Oxford: Clarendon.
Hilbert, D. (1964). „On the Infinite”, in: Philosophy of Mathematics, Englewood
Cliffs, N. J.: Prentice-Hall.
Kanitscheider, B. (1990). „Does physical cosmology transcend the limits
of naturalistic reasoning?”, in: Studies on Mario Bunge’s „Treatise”, (ed.
Weingartner, P. and Doen, G.) Amsterdam: Rodopi.
Swinburne, R. (2004). The Existence of God, Oxford: Clarendon Press.
Anand Jayprakash Vaidya Original Scientific Paper
San Jose State University UDK 164
ON THE ROLE OF MODAL INTUITION

IN MODAL LOGIC
1. Introduction
The philosophy of alethic, as opposed to deontic or epistemic, metaphysical
modality studies the logic, semantics, metaphysical, and epistemic issues
pertaining to modal statements. A statement of alethic modality contains either
the phrase ‘it could have been the case that’ or the phrase ‘ it could not have
been the case that’, or some linguistic variant, such as ‘it is possible that’ or ‘it
is impossible that’. In the philosophy of modality the central goal is to: (a) find
the correct logical codification of alethic modality, (b) the valid rules of modal
reasoning, and (c) the correct formal semantical treatment of modal statements
that does not commit us ontologically to entities or theses about objects and
properties that are unacceptable, given our other commitments. Finally, a
comprehensive treatment includes: (d) an account of how we know or can come
to be justified in believing any number of modal claims.
One of the most important advances in the philosophy of modality was
the introduction of Kripke-style semantics for propositional modal logic by
Saul Kripke. The model he introduced relies on a set of possible worlds W, a
distinguished actual world @, and a binary accessibility relation R defined on
W. It is necessary that P, ‘P’, is true at a world w just in case P is true at every
world accessible from w. It is possible that P, ‘P’, is true at a world w just in case
P is true at some world accessible from w. The following axioms and conditions
characterize various ways in which one can understand the space of possible
worlds via the accessibility relation. The four axioms conjoined constitute the S5
system of propositional modal logic.
K: (P  Q)  (P Q) Normal / Distribution
T: P  P Reflexive: wi[wiRwi]
B1: P  P Symmetric: wiwj[wiRwj wiRwj]
S4: P  P Transitive: wiwjwk[(wiRwj  wjRwk)  (wiRwk)]
The conjunction of the T, B, and S4 axioms can be captured through the
characteristic axiom of the S5 system.
E2: P  P Equivalence (Reflexive, Symmetric, and Transitive)
1 Another way of presenting B is: P  P .

2 Another way of presenting E is: P  P.
Some philosophers, such as Williamson (2013), find the S5 system to be the

most satisfying account of logical and metaphysical modality. However, some do not.
Salmon (1989) argues that the proposition (V) is a counterexample to the S4
axiom of modal logic.
V: A table t, which actually originated from a hunk of matter M carved
from a tree T could have originated from M*, a portion of wood carved
from T that is just slightly different from M.
Dummett (1993) argues that the proposition (U) is a counterexample to the
B axiom of modal logic.
U: There could have been unicorns.
Moving off of the work of Salmon and Dummett: Vaidya (2008) argues
that Salmon’s counterexample to S4 is plausible and can be extended into the
discourse of debates over the epistemology of modality, and Lee Walters (2014)
argues that Dummett’s counterexample to B fails, but the counterexample can be
deployed as a successful counterexample to the E axiom.
In what follows I will develop a version of, what I call, the Dummett-
Walters argument against S53. I will use the Dummett-Walters argument as
well as the Salmon-Vaidya argument to explore an important question in
the philosophy of modality that pertains to the interaction between modal
logic and modal intuition: How should theories of modal logic interact with
potential counterexamples based on modal intuition? I will argue that within the
philosophy of modality there is a tension that borders on paradox when we go
about exploring potential counterexamples to axioms of modal logic. Here is an
introductory sketch of the tension. On the one hand, there are modal intuitions,
intuitions whose content is modal, which can be deployed to challenge certain
axioms of modal logic, such as B and S4. On the other hand, concerning these
modal intuitions one might ask: how do we know that these modal intuitions
are in fact tracking modal reality, as opposed to being mere modal illusions
produced by our own individual modal biases?
The problem can also be stated as follows: if we cannot have any confidence
in a certain range of modal intuitions, why should we trust them in providing
us with reasons to reject or accept specific systems of modal logic? This issue
can be made more problematic by introducing the distinction, also discussed by
van Inwagen (1998), between ordinary and extraordinary modal intuitions. If a
counterexample to an axiom of modal logic rests on an ordinary modal intuition,
then perhaps we have good grounds for rejecting an axiom that is contrary to
the ordinary modal intuition. However, and more problematically: should we be
equally confident in rejecting an axiom of modal logic when the counterexample
it rests on relies on an extraordinary modal intuition?
In exploring the significance of this question I will argue that on two leading
theories, Chalmers’s (2002) conceivability-based account, and Williamson’s
3 It is important to note that Walters’ own clarification of Dummett’s argument does not
involve a commitment to the truth of U, nor to the additional assumptions that are required
for the counterexample to go through. As Walters’ notes at the end of his paper, Dummett’s
argument requires one accepting several claims.
Anand Jayprakash Vaidya: On the role of modal intuition in modal logic 169
(2007) counterfactual-based account, the intuitions against S4 and S5 are highly

problematic. Were the modal intuitions against S4 and S5 known on the basis
of either of these views, they would provide evidence against the views, thus
undermining their own potential role in serving as counterexamples to S4 or S5.
Thus, we are left with the question: how do these intuitions acquire a sufficiently
strong enough epistemic status so as to serve as a counterexample to the axioms
of modal logic that they are deployed against?
In 2 I will develop the Dummett-Walters argument against S5, and bring
it into contact with the Salmon-Vaidya argument against S4. In 3 I will develop
what I call the paradox of modal logic and modal intuition. In 4 I will consider a
response to the paradox based on Goodman’s account of reflective equilibrium
for the case of logic. In particular, I will address a well-known question within
the literature on reflective equilibrium concerning the role of intuitions as inputs.
The question is: Which intuitions matter? With respect to this question I will
attempt to draw a provisional distinction between ordinary and extraordinary
modal intuitions. I will defend a conservative approach to selecting a modal
logic based on modal intuition, on this approach only ordinary modal intuitions
matter. Finally, in 5 I will close by discussing the relevance of restricting the use
of modal intuitions to ordinary modal intuitions as it pertains to Williamson’s
(2014) defense of the counterintuitive thesis of necessitism, on which necessarily
everything is necessarily something.
2. Counterexamples to Axioms of Modal Logic
Dummett-Walters
After explaining Dummett’s purported counterexample to the B axiom and
how Dummett tentatively extends it S4, Walters argues that the counterexample
to B fails, but that it succeeds as a counterexample to S5.
What actually follows [from Dummett’s example] is that S5 is to be rejected.
Let us call the proposition that Dummett is concerned with ‘P’. Dummett claims
that P is necessarily true in u, so that at w we have P. He also claims that
this proposition is possibly false in w, namely, P. But what follows from the
conjunction of P  P is the falsity of E: P  P. (Walters 2014: 4)
Because I am not engaging the question of whether Dummett’s
counterexample is a counterexample to B, I will not engage the excellent work
done by Walters in showing that it is not a counterexample to B. Rather, I will go
on to simply generate the Dummett-Walters argument against S5.
Let w be the actual world.
Let P be the statement: there are unicorns.
Dummett holds, contrary to Kripke, that in w: both P and P hold.
Let u be a world in which there are creatures resembling the unicorns of the
myth, which belong, like deer, to the order Artiodactyla.
Let v be a world in which there are creatures resembling the unicorns of the
myth, which belong, like horses, to the order Perissodactyla.
Necessity of Order Membership, NOM: Where x ranges over species, if x
belongs to order , then x belongs to  essentially, and, since what is essentially
true of x is necessarily true of x, it follows that x.
Let: F = the order Artiodactyla
Let: G = the order Perissodactyla:
From the description of u, v, and NOM, it follows that
 Fx holds in u.
 Gx holds in v.
The Dummett-Walters argument is the following:
1. In the actual world w: P and P are true, since although there are
no unicorns in the actual world, that is P is true in w, it is possible that
there are unicorns and possible that there are not unicorns. In other
words, Dummett holds, contrary to Kripke, that from the actual world
the statement that there could have been unicorns is true and that there
might not have been unicorns is true.
2. P is true in u, since in u there are creatures resembling unicorns,
which are of the order Artiodactyla, and by the necessity of order
membership, being of the order Artiodactyla is essential to all creatures
that are of the order.
3. P is true in w, since u is accessible from w, and in u: P is true.
 
4. P  P is true in w, from (3) and (1) by conjunction.
 
5. P  P is true in w, from (4) and the equivalence of P:: P.

6. [P  P] is true in w, from the equivalence of [  ]:: [  ].
The counterexample can also be captured through the following diagram:
u: F; P
w: P  P Since F and G are incompatible.
v: G: P
Walters articulates in words what is presented in the possible worlds diagram

above. His remarks capture the core point of the argument, as well as the open
question as to which axiom is at fault for the failure.
Now, given that both u and v are accessible from the actual world, w, the
mutual inaccessibility of u and v shows only that the accessibility relation cannot
be both symmetric and transitive, which is to say S5 must go. ... So, if we accept
Dummett’s case, then for all that has been said so far, it could be that transitivity
and S4 are to be given up, rather than symmetry and B. (Walters 2014: 4)
Salmon-Vaidya4
Vaidya (2008) articulates a version of Salmon’s (1989) counterexample to the S4
axiom.
Essentiality of Origin, (EO): If  is the origin of x, then in every possible
world in which x exists, it x is essential, and thus x is true.
The Salmon-Vaidya argument
1. Mt is an actual world fact.
 
2. Mt, from (1) and (EO).
3. V from modal intuition grounded in the view that social kinds, as
opposed to biological kinds, can tolerate vagueness at their origins.
4. Mt, since V grounds M*t.
 
5. Mt, from 4 and  /  equivalence.
 
6. Mt  M. from (2) and (5) by conjunction
 
7. [ Mt  Mt], from (6) and the equivalence of [  ]:: [ 
].
The core idea of the argument comes from V. If the material origin M, which
comes from tree T, of table t can tolerate vagueness, then it follows that although
t has it origin in M essentially, it is possibly possible that t originated from M*, a
portion of T just slightly different from M.
Counterexamples and Questions

The counterexample to S5 offered by the Dummett-Walters argument rests on
U: There could have been unicorns.
4 What is presented here is nothing more than Salmon (1989). The core of the extension
of Salmon’s (1989) argument by Vaidya occurs in deploying it as a potential problem for
Chalmers (2002) conceivability-based theory of the epistemology of modality.
The counterexample to S4 offered by the Salmon-Vaidya argument rests on

In light of V and U, and the fact that S5  S4, we can formulate the following
normative question:
M: Should we reject S5 on the basis of V and U?
V and U are both modal statements. Each is a claim to the effect that
something is possible. U concerns the possibility of a kind of entity, which does
not exist in the actual world, existing in some other possible world. V concerns
the interaction of vagueness with modality with respect to the origins of a material
social kind. Some people have the intuition that both are true. But how seriously
should we take these propositions in thinking about selecting a modal logic? I
believe that we cannot consider M without also engaging the epistemic question:
J: How can we be justified in believing (or gaining knowledge of) V and U?
Moreover, consideration of V and U and the consequences they potentially
have on axioms of modal logic leads us directly to questions concerning the
epistemology of modality. The general view is that the philosophy of modality,
including its semantics, metaphysics, epistemology, and logic, must be confronted
together.
3. The Paradox of Modal Logic and Modal Intuition

The epistemology of modality is concerned with the question: how can
we know or be justified in believing that a modal proposition is true? The
epistemology of modality is thought of as a special area of inquiry because
prima facie it appears that we can know (i) that something that is not actually
true, could have been true, and (ii) something which is true, is also necessarily
true. Knowledge of type (i) and type (ii) prima facie outstrips what knowledge
of the actual world provides us with, since non-actual possibilities are true in
some possible world distinct from the actual world and necessities are true in all
possible worlds. There are many accounts of how we can come to know what is
merely possible or necessary. In recent work, however, there are two important
and well-developed accounts:
Chalmers’s (2002) Modal rationalism, MR.
MR: Ideal Primary Positive Conceivability entails Primary Possibility.
&
Williamson’s (2007) counterfactual-guide, CG.
CG: P  [P c ], If it is possible that P, then it is not the case that were
P true a contradiction would follow.
How would each of these views account for V and U?
On MR the basic idea is that an agent can conceive of a situation in which

V is true, and U is true, and that this conceivability provides at least some
justification for believing that U and V are true. In the case of V, an agent
would conceive of t being present where t does not originate from M, rather it
originates from M*. Since the kind of conceiving involved would be primary, we
would be operating with the primary intensions of t and M and M*, rather than
with their secondary intensions. A static model of the conceived state of affairs
would simply involve an agent reflecting on the proposition t could have come
from M* rather than M. A dynamic model of the conceived state of affairs would
involve an agent running a simulation in their imagination of M* being carved
from T, and then into t. In the case of U, an agent would conceive of animals
resembling unicorns in the relevant respects. But the general idea would be that
by conceiving of a scenario or situation in which U and V are true, we can come
to be justified in believing that U and V are true.
On CG the basic idea is an agent is justified in believing U or V just in case
were they to counterfactually assume the relevant proposition they would not be
led to a contradiction. In the case of V an agent would assume that t originates
from M*, and were they not to detect a contradiction they would be justified in
believing V. In the case of U, an agent would assume that animals resembling
unicorns of the myth are present, and were they not to detect a contradiction
they would be justified in believing U.
Notice, that in both cases, I have not argued that V and U are known or
justified on the basis of either MR or CG. Rather, I have briefly sketched how it is
that one would approach coming to know or acquiring justification for believing
V or U on the basis of these views. It is important to note that if the essentiality
of origins and the essentiality of kind membership are part of what we use to
construct a scenario, in the conceivability case, or to evaluate a counterfactual,
in the imaginability case, we might not be able to claim that there is a scenario
in which V or U are true or that V or U do not lead to a contradiction on the
assumption that they are true. All that has been shown here so far is how we
might go about gaining justification for V and U through conceivability and
counterfactual reasoning in imagination.
Let  stand for either MR or CG, and also connote the tag ‘-theories’
with reference to the two options Chalmers and Williamson provide in the
epistemology of modality. With respect to -theories, the following argument,
called the core argument against -theories, can be formulated.
1.   S5
2. S5
 
3. 
(1) is the main premise that needs to be justified in the case of both MR and
CG. In the case of MR, Vaidya (2008) argues on the basis of MM, as follows.
MM: The space of logical modality is co-extensive with the case of
metaphysical modality.
i) MR  MM
ii) MM  S5
iii) S5

iv) MR
The argument against S5 is of course the Salmon-Vaidya argument against
S4 based on V. In the case of CG a similar argument can be made, except that (i)
and (ii) would have to be replaced by:
(W) CG  S5
However, there is a disanalogy between Chalmers and Williamson with
respect to the core argument. Vaidya (2008) argues for (i), and Chalmers (1999)
explicitly endorses (i). By contrast, Williamson does not explicitly endorse (W)
in the context of the epistemology of modality as developed in Williamson
(2007). Although Williamson (2013) 5 does defend S5, it should be noted that
his mutual endorsement of CG and S5 across (2007) and (2013) could not lead
one to conclude that (W) holds. More importantly, what is crucial is whether the
proof of the logical equivalence in CG requires assuming S5. If the only way to
prove that it is possible that p is logically equivalent to were it the case that p it
would not be the case that a contradiction follows depends on assuming that there
is only one space of possible worlds on which all worlds are accessible to each
other, than there would be reason to hold (W).
In what follows my interest here is in developing a problem that rests on
three claims.
(R) In order to know which system of modal logic, T, B, S4, or S5 is correct
we need to rely on modal intuitions.
(I) Modal intuitions, such as U and V, are inconsistent with -theories.
(D) If -theories are false, then U and V are unknowable.
Explanation of the Problem

Some will object to (R) because they hold that modal intuitions play
no substantial role in identifying which of the many systems of modal logic
is correct. However, it appears that while it is plausible to argue that in some
cases modal intuitions play no role, it appears impossible to hold it in every
case. For example, the system T can be argued to make nothing more than an
analytic claim about ‘necessity’ understood either logically, metaphysically, or
epistemically. The claim is simply that whatever is necessarily true, is true. To
deny this borders on incoherence. Consider (a) and (b):
a) It is necessary that bachelors are unmarried males, but it isn’t true.
b) I believe that Zuleica plays soccer, but it isn’t true.
5 See Chapters 2 and 3 of Williamson (2013) for discussion.

Just as the concept of belief reveals that (b) is prima facie nonsense. The
concept of necessity reveals that (a) is prima facie nonsense. However, the
argument made here cannot be extended to the case of B, S4, and S5. Consider
(c), (d), and (e):
c) It is necessary that bachelors are unmarried males, but it is not
necessarily necessary that bachelors are unmarried males.
d) It is possibly necessary that bachelors are unmarried males, but it isn’t
true that bachelors are unmarried males.
e) It is possibly necessary that bachelors are unmarried males, but it isn’t
necessary that bachelors are unmarried males.
Consideration of (c) – (e) reveals that one must go beyond mere conceptual
truths or direct understanding. I hold that while (b) rests on a conceptual truth
about necessity, (c), (d), and (e) do not. As a consequence knowledge of B, S4, and
S5 requires consideration of the metaphysics of modality, some of which rests on
modal intuitions concerning kinds or a substantial scientific theory of kinds.
(I) is also controversial. Roca-Royes (2006) has argued that one can use a
modified version of Peacocke’s Unified Modal Extension Principle to show that
V can be made consistent with S4. The basic idea is that rather than holding that
M is the piece of wood from which t originates, we simply hold that M1....Mn is
the range of pieces of wood from which t originates. As a consequence, what is
necessary is that t originates from the range M1...Mn and not M. That is rather
than let V stand as a counterexample to S4, make it the case that because social
kinds can tolerate flexibility at their origin, the Unified Modal Extension Principle
requires us to select the range, instead of the particular slice in the range that is
M. Roca-Royes’s proposal is quite engaging and powerful. However, it would only
apply to the Salmon-Vaidya argument, and not to the case of the Dummett-Walters
argument. So, (I) is surely controversial. But when we push harder on this case
what is revealed is that there are two options: either deny the intuition concerning
the flexibility of origins or deny the inconsistency of the intuition with S4.
(D) is not as controversial as either (R) or (I). Rather, what (D) brings to
the forefront is the epistemic question about justification. More importantly,
it invites us to start inquiring into what other options outside of -theories
are available for coming to know modal truths. Let me close this section by
considering another option, which I don’t believe gets us out of the problem.
Consider Lowe’s (2012) proposal for how we come to know modal truths.
On his account we come to know what is necessary by way of essence. This sets
up a deductive model for arriving at knowledge of possibility.
A knows that E is the essence of x.
A knows that if E is the essence of x, then Ex.
A knows that F is inconsistent with E.
If F is inconsistent with E, then Fx.

Fx.
There is something compelling about this strategy that makes it attractive in

contrast to conceivability-based reasoning or counterfactual reasoning. The basic
idea is that we know that U and V are true simply by first coming to know the
essence of the relevant objects involved, such as unicorns and tables. Once we
know the relevant essential properties we are in a position to deduce whether U
or V are true. However, as is clear by the nature of the approach, this setup pushes
the question back further. Rather than asking how we know what the modal
properties of an object are we are now asking: what are the essential properties
of the object? However, we face two problems on this approach. First, it could be
that in all cases essence grounds modality, but in some cases we come to know
about modality through some route other than through deduction from essence.
Second, even if we do know about modality through essence: How do we sort
the essential properties from the merely accidental? Arguably, the latter question is
just as difficult as the question: how do we sort the possible from the necessary.
Lowe argues that knowledge of modality does not depend on imagination,
conceivability, or intuition. But then what does it depend on? He argues that
it depends on grasp of essence. But this just leaves us in the dark on the whole
matter. One might simply concede the structural point: knowledge of modality
depends on knowledge of essence, but then argue that intuition in conjunction
with knowledge of real definitions gives us knowledge of essence. Why should we
trust our intuitions of essence? And: Which intuitions should we trust?
4. Reflective Equilibrium and the

Conservative Approach to Intuition
A natural answer to both of the questions raised in the last part of the
previous section, and concerning the whole project of selecting the correct modal
logic, comes from Nelson Goodman (1955). He famously says the following
about how the rules of deductive inference are justified:
I have said that deductive inferences are justified by their conformity to
valid general rules, and that general rules are justified by their conformity to
valid inferences. But this circle is a virtuous one. A rule is amended if it yields an
inference we are unwilling to accept; an inference is rejected if it violates a rule we
are unwilling to amend. The process of justification is the delicate one of making
mutual adjustments between rules and accepted inferences; and in the agreement
thus achieved lies the only justification needed for either. (Goodman 1955: 67)
Surely if this method works in the case of deductive inference, it can be
made to work in the case of identifying the correct modal logic. For example, K
and T seem to be undeniable, while B, S4, and S5 are all questionable. We would
hope that the identification of the correct modal logic could be achieved on the
basis of the greatest coherence between our modal intuitions and other principles
of logic. But this pushes out the question that really needs to be addressed, and
which has been looming since the beginning: Which modal intuitions should
we trust? Skeptics would say none. Liberals would say all. But I think a more
promising approach is available between the skeptical and the liberal position.
That approach takes us to a conservative picture of which intuitions should form
the data from which we attempt to identify the correct modal logic.
Let me begin articulating the conservative approach through a rough
distinction between two kinds of modal claims.6
Ordinary modal claims include the following:
i) It is possible for an unbroken table to be broken.
ii) A chair located in one part of a room could be located at another part
of the room.
iii) A person born as a non-twin could have had a twin.
iv) A table t* carved from M*, distinct from t carved from M, could have
been made from T.
Extraordinary modal claims include the following:
v) There could have been unicorns.
vi) Zombies (creatures physically identical to humans, but lacking
consciousness) could have existed.
vii) There could be a perfect duplicate of f.
viii) A table t which is carved from a hunk of wood M taken from tree T
could have been carved from a hunk of matter M* just slightly different
from M and still been the very same table.
Now the jarring, but crucial part: No precise distinction between the
ordinary and the extraordinary can be drawn. To seek an account between the
two with exact necessary and sufficient conditions is to engage in a philosophical
folly. However, some comments about the items on both lists can be made, and
also some structural features of the distinction can be noted. I will begin with
a comparative commentary, and then move into discussion of the structural
features.
Notice that (i) and (ii) concern ordinary objects in ordinary situations. They
also are in the penumbra of modal judgments we often make. We think about
whether a table can break and under what circumstances it might break when
purchasing one. And likewise we think about where a chair could be located in a
room relative to other furniture when we are thinking about arranging things in
a room. By contrast, (v) and (vi) are statements that, if we believe to be possible,
are likely based on stories that contain descriptions of such creatures. While
these stories can involve highly creative descriptions, they are often not based
on, nor contain the kinds of descriptions that would be necessary for one to have
even a reasonable belief in the existence of such things. The point can be more
carefully attended to by noting that within the genre of science fiction it is well
known that the stories are successful when they have the right blend of fantasy
and science.
6 There are many papers in which the distinction between ordinary and extraordinary (perhaps
the use of different words) is appealed to. Van Inwagen (1998) serves as one historically
important place, more recently, Bueno & Shalkowski (2014) serves as another place.
Notice that (iii) and (vii) look similar with respect to duplication. However,
the claim in (iii) is quite innocuous and in the realm of the ordinary, while
(viii) appearing innocuous, is actually unclear. In (iii) we know what processes
account for a twin to be born. In (vii) we don’t know what it would mean for f to
have a perfect duplicate by way of a process. We know what perfect duplicate is,
but we don’t know what process could bring it about.
Notice that (iv) and (viii) also look similar. But again there is a key difference.
In (iv) the claim is that another table looking similar to the original table could
have been made from matter just slightly different from the matter that was used
to create the original table. The claim in (viii) is about one and the same table
having a modal property that of being possibly created from matter distinct from
that from which it was originally created. The distinction is that (iv) is what could
be created from the tree T by way of a process, while (viii) is about a process and
identity. Moreover, (viii) is stronger than (iv).
Now that we have some details about the list down, we can also note
some structural features of the distinction. First, and foremost, the distinction
between ordinary and extraordinary modal claims changes over time as we
learn more about the actual world both in science and mathematics. What is
legitimately believed to be possible or impossible at one time can later be deemed
impossible or possible simply through consideration of the relevant science and
mathematics. This meta-modal-epistemic insight is grounded in experience of
the actual world as discoveries are made in mathematics and science. For brevity
I mention only two examples.
At one period of time in mathematics it was thought that there is only one
size of infinity. Then G. Cantor’s argued that the real numbers could not be put
into a one-to-one correspondence with the natural numbers. At one period of
time in physics it was thought that the speed at which an object was moving
could not effect time. Then A. Einstein argued that an object moving close to
speed of light would slow time.
Second, the distinction between the two groups does not rest on the
phenomenology of reflection. It is not that ordinary modal propositions feel
more certain on reflection than extraordinary modal propositions. For example,
consider the counterpossible: if the number 2 were the number 3, then ‘2 + 2
= 6’ would be true. The antecedent of the conditional is false in every possible
world, since it is impossible for 2 to be 3. The phenomenology of reflection on
the truth of the counterpossible is as strong as the phenomenology of reflection
on the truth of the unrealized possibility: although Miranda does not have a
brother, she could have had a brother. However, the closest possible world in
which the counterpossible is true can be argued to be much farther away from
the actual world as the unrealized possibility claim. Thus, some extraordinary
modal claims may have the same phenomenal reflection strength as simple
ordinary modal claims about unrealized possibilities. In addition, although more
controversial, some modal theorists might argue that the claim that unicorns
could have existed is as phenomenologically certain as the claim that a table that
is unbroken could have been broken.
Third, ordinary propositions do not distinguish themselves from the
extraordinary in virtue of how distant they are from the actual world. For the very
notion that w1 is more distant from @ than w2 is riddled with thorny problems,
two of which include the problems caused by the existence of impossible
worlds and the relativity of the similarity relation across worlds. In addition,
the counterpossible discussed above, is potentially case of a proposition that is
distant from the actual world, yet knowable.7
Now if the distinction is granted, then the approach to selecting a modal
logic I believe we should adopt is the conservative approach, CA.
CA: Ordinary modal propositions form the data from which a modal logic
should be brought into alignment through reflective equilibrium, and
not extraordinary modal claims.
CA has two primary benefits. First, while it is conservative as to which
data it allows in, the approach it takes to which modal logic we are justified
in believing in at a given time, it is also permissive with respect to change.
Were an extraordinary modal proposition to find additional support it would
then become an ordinary modal proposition, which in turn could force an
adjustment as to which modal logic we should accept. Second, it has the benefit
of focusing our attention on understanding theories, which in turn have modal
consequences, rather, than focusing on modal intuitions based merely on their
phenomenology. The second benefit does not aim to downplay the importance
of the phenomenology of modal intuitions. For the phenomenology of modal
intuitions has a role to play in an account of which modal logic we are to accept.8
Moreover, the focus of the second point is to push our attention to other areas
of mathematical and scientific inquiry that may corroborate our intuition. Just
as science and mathematics are blind without metaphysics, metaphysics without
science and mathematics is empty. As a consequence, the debate over the
significance of Dummett’s argument and Salmon’s argument is that they rest
on extraordinary intuitions, intuitions whose weight is significantly weak with
respect to the project of determining the correct modal logic. More weight ought
to be given to theoretical conditions about modal logic. One key exemplification
of this weighting issue between intuition and theoretical considerations is tackled
by Williamson (2013) in his articulation and defense of necessitism.
5. The Debate over Necessitism

Williamson (2013) defines necessitism and contingentism as follows:
Call the proposition that it is necessary what there is necessitisim, and its
negation contingentism. In slightly less compressed form, necessitism says that
7 It would be inappropriate of me not to mention the excellent work of Kment (2014) on

defining the similarity relation across worlds and discussing how comparison and weighting
can be done across worlds. Nevertheless, we cannot conclude from the fact that an intuitive
modal proposition is a close modal proposition, nor that a distant modal proposition is a
counterintuitive proposition.
8 For an excellent theory of how the phenomenology of intuition plays a role in justifying our
beliefs see E. Chudnoff ’s perceptual theory as articulated and defend in his (2013).
necessarily everything is necessarily something; still more long-windedly: it is

necessary that everything is such that it is necessary that something is identical
with it. In slogan: ontology is necessary. Contingentism denies that necessarily
everything is something. In a slogan: ontology is contingent. (Williamson 2014:
2)
As an example of a proposition that necessitism and contingentism will
differ over, Williamson give the example of (C).
C: This coin, which actually exists now, could have failed to exist.
According to contingentism (C) is true, because what there actually is
could have failed to be. According to necessitism (C) is false. Taking note of the
obvious jarring content of necessitism, Williamson raises an obvious question
about it with a corresponding argument for a specific conclusion.
Why take necessitism seriously? Isn’t it just obvious that many things, such as
the coin, could have not been? [...] If a philosopher produces a clever theoretical
argument for necessitism, we may learn much from diagnosing the fallacies in it,
just as we may learn much from diagnosing the fallacies in a clever argument for
radical skepticism, but in each case are we not entitled to confidence in advance
that the argument will indeed be fallacious. (Williamson 2014: 5).
In the prior section I argued that we should not take extraordinary intuitions
about modal maters seriously in selecting a system of modal logic. How does
that claim interact with the prospects of Williamson’s extended defense of the,
admittedly, counter-intuitive thesis of necessitism? Contrary to how things
might seem, the claim that we should only consider ordinary modal intuitions
does not yield even a prima facie argument against necessitism. To draw out the
reason why consider (C) alongside (N):
(C): This coin, which actually does exit, could have failed to exist.
(N): This coin, which actually does exist, could not have failed to exist.
Is (C) somehow more intuitive as a modal proposition than (N)? The
question generally attracts the answer that (C) is far more intuitive than (N). On
the basis of the greater attraction of (C) over (N) some may be inclined to argue
as follows:
1. Intuitive modal propositions are ordinary modal propositions.
2. Ordinary modal propositions must be considered in investigating
systems of modal logic.
3. (C) is an intuitive modal proposition.
4. So, (C) should be taken seriously in investigating systems of modal
logic, and since (C) is inconsistent with necessitism, in advance of
considering the arguments for it, we can be confident that it should be
rejected.
By unpacking the argument this way I believe we can expose a cognitive
blind spot in our modal theorizing. (1) is false. The intuitive and the ordinary do
not over lap. Some intuitive propositions are non-ordinary, even if they fall short
of being extraordinary. Some unintuitive propositions are ordinary, even though
they are unintuitive. Since our concern is with the unintuitive that might turn
out to be ordinary, consider an example from mathematics, the Banach-Tarski
Theorem (BTT).
(BTT): Given a solid ball in 3-dimensional space, there exists a decomposition
of the ball into a finite number of non-overlapping pieces (i.e. disjoint
subsets), which can then be put back together in different ways to yield
two identical copies of the original ball.
For the purpose of simplification, and to make explicit the modal claim, we
can reduce (BTT) to (B).
(B): Where B is a solid ball in 3-dimensional space it is possible to convert B
into two identical copies of itself.
(B) is counterintuitive, but it is not extraordinary. While the content of the
claim is counterintuitive, the existence of a mathematical proof of it puts it in
the realm of ordinary modal propositions. Ordinary modal propositions track
established scientific and mathematical theories and propositions deducible
from them. While it is true that (N) is unintuitive, it is no more unintuitive
than (C). What is relevant is whether there are scientific and mathematical
facts that also play a role in justifying our belief in either (C) or (N). Take the
analogy with (B) again. (B) is counterintuitive, but this does not mean we should
reject it. Rather, because we also have a proof of it, we should disregard the
counterintuitive nature of (B) and seek another explanation for why we think
that (B) is counterintuitive. It could be that (B) is like a Müller-Lyer illusion,
even though we believe the lines to be the same length we cannot but seem them
as being of different lengths. Likewise, (N) is counterintuitive, but in light of the
sustained arguments that Williamson has offered in defense of (N), it maybe
that we have to look at (N) as a counterintuitive claim that rests on significant
logical and metaphysical theorizing and proof. Our continued belief that (N) is
counterintuitive may be nothing more than an illusion similar to what is at play
in the case of (B) and the lines of a Müller-Lyer diagram.
References
Chalmers, D. 1999. Materialism and the Metaphysics of Modality. Philosophy
and Phenomenological Research 59: 473–496.
----------------. 2002. Does Conceivability Entail Possibility? In T. Gendler and
J. Hawthorne (Eds.) Conceivability and Possibility. Oxford University Press.
Chudnoff, E. 2013. Intuition. Oxford University Press.
Dummett, 1993. Could There Be Unicorns? In his The Seas of Language. Oxford
University Press: 328–348.
Goodman, N. 1955. Fact, Fiction and Forecast. Harvard University Press.
Kment, B. 2014. Modality and Explanatory Reasoning. Oxford University Press.

Lowe, E. J. 2012. What is the Source of Our Knowledge of Modal Truths? Mind
121. 484: 919–950.
Roca-Royes, S. 2006. Peacocke’s Principle-Based Account of Modality: „Flexibility
of Origins” Plus S4. Erkenntins 65: 405–426.
Salmon, N. 1989. The Logic of What Might Have Been. The Philosophical Review
98: 3–34.
Salmon, N. 1993. This Side of Paradox. Philosophical Topics 21: 187–197.
Vaidya, A. 2008. Modal Rationalism and Modal Monism. Erkenntins 68: 191–212.
Van Inwagen, P. 1998. Modal Epistemology. Philosophical Studies 92: 67–84.
Walters, L. 2014. The Possibility of Unicorns and Modal Logic. Analytic Philosophy
Williamson, T. 1990. Identity and Discrimination. Blackwell Publishers.
------------------. 1998. Bare Possibilia. Erkenntnis 48: 257–273.
------------------. 2007. The Philosophy of Philosophy. Oxford: Wiley-Blackwell.
------------------. 2013: Modal Logic as Metaphysics. Oxford: Oxford University
Press.
CIP – Каталогизација у публикацији
Народна библиотека Србије, Београд
1
FILOZOFSKI godišnjak /
Belgrade Philosophical Annual: glasnik
Instituta za filozofiju Filozofskog fakulteta u
Beogradu / glavni i odgovorni urednik Slobodan
Perović. – God 1, br. 1 (1988) – God. 27, br. 27 (2014). –
Beograd (Čika Ljubina 18–20): Institut za
filozofiju Filozofskog fakulteta, 1988-
(Beograd: Službeni Glasnik). – 24 cm
Godišnje
ISSN 0353–3891 = Filozofski godišnjak / Belgrade Philo-
sophical Annual
COBISS.SR-ID 15073792

BPH A

Enviado por

Dados do documento

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

BPH A

Enviado por

Direitos autorais:

Formatos disponíveis

BELGRADE PHILOSOPHICAL ANNUAL

/FILOZOFSKI GODIŠNJAK 27/2014

Belgrade Philosophical Annual online at http://www.f.bg.ac.rs/bpa/index.html

EDITOR’S NOTICE .............................................................................................. 5

Belgrade Philosophical Annual is published by the Institute for Philosophy,

Timothy Williamson Original Scientific Paper

HOW DID WE GET HERE FROM THERE? THE

Opponents of analytic philosophy often associate it with logical positivism.

How did a species of philosophy with so much anti-metaphysics in its gene

modal operators such as ‘possibly’ and ‘necessarily’. He also compared them

logic. Quine was especially critical of quantified modal logic, as developed by

Writing in 1981, Lewis described ‘a reasonable goal for a philosopher’ as

central core. In Strawson’s view, that structure is not hierarchical, as envisaged

Word and Object, Quine emphasized the advantages in logical smoothness of

ways of thinking, so not radical in relation to common sense. Of course, the

ordinary language philosophers, even when ordinary language plays no special

least as much a philosopher of language as a metaphysican in the 1970s), and

Davidsonians did not expect the philosophy of language to be independent

the cash-value of alternative metaphysical pictures. Methodologically, he proposed

Another memory comes from a supervision with Dummett. We were

fifty were more sympathetic to Hurley’s objections. The second snapshot is of

semantics, like that associated with Russell, in explaining linguistic phenomena

else, or that no other discipline uses them at all; nevertheless, philosophical

Quine, Willard van Orman 1951: ‘Two dogmas of empiricism’, Philosophical

Jürg Steiner Original Scientific Paper

THE DYNAMICS OF DELIBERATION

This paper is a continuation of previous research on the deliberative model

we could explain variation in these aggregated values in their antecedents and

Although we do not look for authoritative stories, we are still interested

If the issue of corruption is sufficiently linked to the peace process the

12 Simon Niemeyer, Deliberative Monetary Valuation as a Political Economic Methodology:

Citizens’ Parliament actually talks about a „transformative” incident, when

20 Johan Galtung. Conflict Transformation By Peaceful Means: A Trainer’s Manual. United

DEMOCRACY AND TRUTH:

Independent Objectivity as a Target:

First Aspect: Scientific Truth and Democratic Society

Second Aspect: Solving a Practical Social Problem--

measures of progressive taxation. We wouldn’t be very surprised if they failed as

Deliberation as a Possible Way-Out

establishes the standards of „communicative action”,12 which excludes threats,

What structure does a group need to have in order to reach

solve the problem of technicalities associated with deliberative undertaking on

3d Aspect: Creating Autonomous Citizens – the

„Machiavellian” Features of a Government20

Transparency is important in any democracy, and especially to deliberative

DELIBERATIVE DEMOCRACY AND THE

One of the central assumptions of deliberative democratic theory is that

mechanical information-pooling and discursive information-pooling. You are

It seems to me that a deliberative democrat faces the following dilemma.

I take this difference between us as a relevant difference, one which effectively

view of conformists, despite an agreement with many other people, we should

Democracy: Evidence from Deliberative Polls,” PSPE working papers, 01–

Jan Narveson Original Scientific Paper

Abstract: I will be arguing that democracy as classically and normally understood

Democracy is almost incredibly popular among social theorists and various

There are two, though related, general problems about democracy.

1. The Constituency Problem

2. The Agenda Problem

And so a decent „democratic” society will have constitutional restrictions

We have, in other words, a mechanism for making decisions that is obviously

A way of putting this difficulty is to note that the constituency of different

This problem is perfectly general. No matter where in the spectrum our

Public Goods, again