doi:10.1016/j.shpsa.2008.03.013


Stud. Hist. Phil. Sci. 39 (2008) 290–293
Contents lists available at ScienceDirect

Stud. Hist. Phil. Sci.

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / s h p s a
Essay Review

Conditions may apply

Paul Dicken
Churchill College, University of Cambridge, Storey’s Way, Cambridge CB3 0DS, UK
When citing this paper, please use the full journal title Studies in History and Philosophy of Sciences
Every thing must go: metaphysics naturalized
James Ladyman & Don Ross (with David Spurrett & John Col-

lier); Clarendon Press, Oxford, 2007, pp. x+345, Price £45.00 hard-
back; ISBN 978-0-19-927619-6.

Every thing must go is an exercise in what James Ladyman and
Don Ross (with substantial contribution from David Spurrett and
John Collier) call ‘naturalistic metaphysics’. Negatively construed,
this consists in the systematic rejection of what the author’s re-
gard as overly scholastic theorising; the marks of the modern
day schoolman being a tendency to ignore the relevant science,
using outdated or ‘domesticated’ science rather than cutting edge
research, and an implausible (not to say, arrogant) faith in the
powers of their own a priori intuitions (p. 17). David Lewis, with
his cost-benefit approach to ontological commitment, and his Hu-
mean Mosaic of ‘perfectly natural properties of points, or point-
sized occupants of points’ (Lewis, 1999, p. 226) is a prime exam-
ple of just such a philosophical villain, committed as he was to
systematising his own intuitions without due deference to scien-
tific authority, and labouring as he did under a long since discred-
ited atomistic conception of the physical world. Kim’s (1998)
defence of physicalism, van Inwagen’s (1990) ontological atom-
ism, and Armstrong’s (1983) doctrine that ‘everything that exists
is in space and time, despite the fact that contemporary physics
takes seriously the idea that space-time itself is emergent from
some more fundamental structure’ (Ladyman et al., 2007, p. 23)
also feature prominently in the naturalistic metaphysician’s rogue
gallery. By contrast, Ladyman et al. ‘are not concerned with pre-
serving intuitions at all, and argue for the wholescale abandon-
ment of those associated with the image of the world as
composed of little things, and indeed of the more basic intuition
that there must be something of which the world is made’ (p. 12),
and go on to ‘deny that a priori inquiry can reveal what is meta-
physically possible’ (p. 16), citing non-Euclidean geometry, non-
deterministic causation and non-absolute time as possibilities ru-
led out by philosophical reflection, but ultimately established by
scientific research.
0039-3681/$ - see front matter � 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.shpsa.2008.03.013

E-mail address: ped21@cam.ac.uk
More positively, the programme is a commitment to take sci-
ence seriously; so seriously in fact that the only legitimate role left
for a thoroughly naturalised metaphysics is to provide an (empir-
ically well confirmed) unification of existing scientific theorising.
This is captured in what Ladyman et al. refer to as the Principle
of Naturalistic Closure, which states that:

Any new metaphysical claim that is to be taken seriously at
time t should be motivated by, and only by, the service it would
perform, if true, in showing how two or more specific scientific
hypotheses, at least one of which is drawn from fundamental
physics, jointly explain more than the sum of what is explained
by the two hypotheses taken separately. (p. 37)

where a hypothesis is considered ‘scientific’ on grounds of institu-
tional consensus, potential for bona fide investigation/confirma-
tion, and is pursued for the sake of truth rather than mere
utility. This methodology is further constrained by the Primacy
of Physics Constraint, which specifies that:

Special science hypotheses that conflict with fundamental
physics, or such consensus as there is in fundamental physics,
should be rejected for that reason alone. Fundamental physical
hypotheses are not symmetrically hostage to the conclusions of
the special sciences. (p. 44)

Ladyman et al. assure us that the Primacy of Physics Constraint ‘is
a regulative principle in current science, and should be respected
by naturalistic metaphysicians’ and that ‘the first, descriptive,
claim is reason for the second, normative, one’ (ibid.).

The overall result is ontic structural realism—the view that
there are no things, only (objective, modal) structure—a position
that has been developed by French and Ladyman over the past sev-
eral years in a number of papers (e.g. French, 1989, 1998; French
and Ladyman, 1997, 2003; Ladyman, 1998). More concretely, ontic
structural realism is explicitly motivated from two directions:
firstly, as the most plausible reconciliation of competing intuitions
over both the no-miracles argument and the pessimistic meta-

mailto:ped21@cam.ac.uk
http://www.sciencedirect.com/science/journal/00393681
http://www.elsevier.com/locate/shpsa


P. Dicken / Stud. Hist. Phil. Sci. 39 (2008) 290–293 291
induction, and moreover, as the dialectical synthesis of traditional
scientific realism and constructive empiricism (Chapter 2); and
secondly, as the correct metaphysical picture forced upon us by
considerations in the philosophy of physics—interpretative issues
in quantum mechanics and the general theory of relativity here
taking centre stage (Chapter 3).

The second half of the book is subsequently devoted to the
promised project of scientific unification, and the attempt to show
how an ontology consisting solely of structure-without-individuals
can be made consistent with a naturalistic approach to the special
sciences. The problem here of course is that the special sciences
appear to deal heavily in individuals: the naturalistic metaphysi-
cian therefore faces the dilemma of either treating the special sci-
ences instrumentally (i.e. that they are mistaken over their own
ontology) and thus violating his own naturalistic constraint, or
abandoning unification in favour of a fundamentally disordered
world (à la Dupré, 1993) and thus failing to provide the proposed
unification, one of the central desiderata of his project (p. 196).

The solution (Chapter 4) is to construe the ‘individuals’ that fea-
ture in the special sciences instrumentally, but to show how they
are supported and legitimised by objective structures correctly
identified and investigated by the discipline in question; as Lady-
man et al. put it ‘individual objects, events, and properties are de-
vices used by observers (when these observers aren’t making
mistakes) to keep cognitive books on what science finds to be
sufficiently stable to be worth measuring over time, viz. some
but not other patterns’ (pp. 228–229). The appropriate patterns
are ‘real patterns’, in the sense originally proposed by Dennett
(1991)—counterfactually robust and providing a suitable degree
of informational compression—and suitably elaborated, in infor-
mation-theoretic terms, to be fully compatible with the Principle
of Naturalistic Closure. It follows then that since ‘some of these
patterns are indeed conceptualised as individuals in some special
sciences, while simultaneously not being so conceptualised by
other special sciences making projections at other scales of resolu-
tion’ (p. 229), Ladyman et al. are committed to what they call the
scale-relativity of ontology; or in the phrase originally coined by
Ross (2000), ‘rainforest realism’. Finally (Chapter 5), an account
of causation is developed along similar naturalistic lines—as a
heuristic for tracking real patterns—that legitimises its role in the
special sciences, yet also respects the Russellian suspicion over
the existence of such processes in fundamental physics.

The book is both engagingly pugnacious and clearly argued; it
covers a broad range of topics from the philosophy of science,
the philosophy of physics and the philosophy of the special sci-
ences, and although a passing familiarity with a large number of
often technical debates is frequently assumed, it remains com-
mendably accessible. The overall position is also quite compelling:
although many will still harbour substantial misgivings over the
central metaphysical notion of a relation sans relata (pp. 154–
156), it is in terms of the overall attractiveness of the position,
and in particular its ability to reconcile and resolve outstanding de-
bates within the philosophy of science, that Ladyman et al. ulti-
mately stake their case.

In particular, the invective against so-called neo-scholastic
metaphysics seems well founded: if you’re going to appeal to ‘con-
temporary science’ as a major premise in your philosophical argu-
ment, you should at least get the science right. The same principle
presumably holds for the history of science too; here however
Ladyman et al. are on shakier ground. As with all structuralists,
their solution to the pessimistic meta-induction is hostage to
whether or not successive scientific theories really do display sub-
stantive structural continuity (pp. 93–95). A handful of references
are duly supplied to this effect: Worrall’s (1989) well known ac-
count of the structural continuity between Fresnel’s ether theory
and Maxwell’s electromagnetic field; Saunders (1993) on the
structural continuity between Ptolemaic and Copernican astron-
omy; Brown’s (1993) study of classical and relativistic mechanics;
Bain and Norton (2001) on descriptions of the electron; and Lyre
(2004) on Maxwellian and quantum electrodynamics. Yet as with
most philosopher’s histories, these are accounts that abstract from
any kind of realistic historical context. And without further argu-
ment, it is far from clear how these retrospective recoveries of pre-
vious structure (invariably only as very special limiting cases) is
supposed to constitute a demonstration of structural preservation.
Moreover, any such ‘preservation’ must also be shown to be tran-
sitive if it is to provide any support for the structuralist’s case,
something the current piecemeal approach goes nowhere near
addressing. In response to the challenge that sometimes structure
is actually lost in theory-change (e.g. Stanford, 2003, pp. 570–572),
Ladyman et al. are content to note that ‘the problem is surely not
analogous to the one the [traditional, full blown] realist faces with
respect to ontological discontinuity’ (p. 157), but otherwise dis-
miss the issue. To be fair, the authors are not committed to some-
thing as implausible as formal/syntactic continuity: mathematical
structure is merely a representation of the real, ontic structure. But
without clearer criteria for what exactly is supposed to be pre-
served on this account, and how this representational relation is
supposed to be construed, one begins to wonder what it would
be for theory-change to fail to preserve structure.

A more substantial worry concerns the naturalistic agenda of
the project. Constraining one’s metaphysical ruminations in accor-
dance with the Principle of Naturalistic Closure and the Primacy of
Physics Constraint seems both well–motivated and in many cases
eminently sensible. However, there are two substantial issues here
which Ladyman et al. fail to address. The first is the extent to
which background metaphysical assumptions may in fact be an
important impetus to scientific research. This may be the case
when super-empirical assumptions suggest and make plausible
new avenues of scientific investigation (consider the often explic-
itly philosophical agendas at play in early debates over the com-
pleteness of quantum mechanics), or when such presuppositions
help provide the sort of institutional conservatism that Kuhn ar-
gued is necessary for effective ‘normal science’. By stipulating that
the relationship between metaphysics and science be asymmetri-
cally constrained by the Primacy of Physics Constraint, Ladyman
et al. risk constructing their philosophical position at the expense
of an impoverished account of actual scientific practice.

My first worry then is that the pursuit of a thoroughly natural-
istic metaphysics threatens to reject too much philosophical pre-
supposition. My second worry is that such a pursuit also
threatens to reject too little philosophical presupposition, and is
based on the simple observation that many contemporary scien-
tific theories are themselves ‘neo-scholastic’ insofar as they contain
(naturalistically unjustified) metaphysical assumptions. Take for
example the special theory of relativity (SR), and the philosophical
debate over the nature of time to which it relates. It is sometimes
argued that since SR entails the observer-relativity of the present
moment, any irreducibly tensed theory of time must be abandoned
(Putnam, 1967, is a classic example of this kind of argument).
However, the core components of SR—the equivalence of inertial-
frames and the constant speed of light in a vacuum—only entail
the relativity of the present in conjunction with a specific definition
of simultaneity, which in its traditional formulation is an explicit
expression of Einstein’s background verificationist assumptions.
Bourne (2006, pp. 160–186) argues this point at length; as he puts
it, this definition ‘is an operational definition: it is a reductive anal-
ysis of simultaneity in terms of observations of light signals under
certain specified conditions, and so satisfies [Einstein’s verifica-
tionism]. But operationalism is untenable, and any independent
support it has derives from accepting [verificationism], which
has been rejected’ (ibid., p. 172). The point then is that this is a



292 P. Dicken / Stud. Hist. Phil. Sci. 39 (2008) 290–293
component of SR that one can reject without doing damage to the
theory proper (as Bourne subsequently goes on to do) and thus SR
is not incompatible with a tensed theory of time. Although Lady-
man et al. ultimately draw no metaphysical conclusions from SR,
on the basis that it is only ‘a partial physical theory that cannot de-
scribe non-inertial frames of reference, or gravity’ (p. 164), they do
explicitly reject any attempt to reconcile a tensed theory of time
with SR in the manner outlined above on the grounds that ‘The
[Principle of Naturalist Closure] forbids the revision of scientific
theories on purely philosophical grounds, so the proposal of a priv-
ileged [present] contra SR requires a scientific motivation’ (ibid., n.
61). Presumably however, one does not need a scientific motiva-
tion to overhaul someone else’s fossilized (not to mention, long
since discredited) metaphysical assumptions. Yet without a more
discriminating conception of the properly empirical components
of a scientific theory, this is exactly what the afore-mentioned
principle commits us to: naturalistic metaphysics therefore threat-
ens to reduce to conservative metaphysics.

A similar hint of the metaphysical status quo can be found in the
alleged synthesis of scientific realism and constructive empiricism
that provides the first substantive argument for ontic structural
realism. According to Ladyman et al., the analytical balancing act
forced upon us by the competing intuitions of the no-miracles
argument and the pessimistic meta-induction shows us that full
blown realism is unsustainable.1 The natural retreat is to a more
selective attitude of scientific belief, as for example advocated by
van Fraassen. Yet constructive empiricism is also unsustainable
insofar as it dispenses with an objective account of modality. The
problem here—as originally developed by Ladyman (2000)—is that
the constructive empiricist restricts his beliefs to the observable
consequences of his accepted scientific theories, a modal distinc-
tion that stretches to encompass what we might have observed,
had the circumstances been different. Yet according to van Fraas-
sen, there are no objective modal facts: something counts as obser-
vable (roughly speaking) insofar as it can be represented in the
right way in our scientific models, an ultimately subjective and
conventional fact about the scientific community. But this makes
the distinction between observable and unobservable entities too
arbitrary to do the metaphysical heavy-lifting required of it. The
constructive empiricist should presumably then adopt a more
objective account of modality; yet this violates his basic epistemic
principles, and collapses the position into a form of structural
realism.

Unsurprisingly however, van Fraassen disagrees with this diag-
nosis. In their reply to Ladyman, Monton and van Fraassen (2003)
argue that their proposed metalinguistic account of modality is far
from arbitrary: the truth-conditions for their account may indeed
turn on conventional decisions of the scientific community, but
this will itself be constrained (and, arguably, justified) by the de-
mands for empirical adequacy. The basic point however is that
for empiricists who deny the notion of objective modality alto-
gether, to demand with Ladyman et al. that only an objective ac-
count of modality is good enough to ground their distinction is
simply to beg the question against them. Moreover, even if one
sympathises with the objection mounted here, it must be conceded
that the demand for a more robust account of modality on the part
of the constructive empiricist is a purely metaphysical demand: the
extent to which one finds the metalinguistic approach to counter-
factuals advocated by Monton and van Fraassen unsatisfactory will
be determined by the sort of pre-scientific cost-benefit analysis of
ontological commitment that one brings to the debate, the sort of
thing championed by Lewis and supposedly renounced by the nat-
1 Ladyman et al. refreshingly give shortshrift to any realist appeal to an increasingly baroq
is really how it is described to be by scientific theories is not an issue in the philosophy of la
that terms like ‘‘ether” refer, there is no doubt that there is no elastic solid permeating al
uralistic metaphysician. Indeed, a purely philosophical debate over
the attractive features of a theory of modality is something that
floats quite freely of any naturalistic constraint; and no debate
over the appropriate semantics for modal statements is going to
make any headway in a unified, naturalistic metaphysics.

A similar tension—between the pursuit of a thoroughly natural-
istic agenda and the ineliminable appeal to metaphysical presup-
positions in defending their version of it—can also be found in
the chapter on the philosophy of physics, where Ladyman et al.
provide their strongest case for ontic structural realism. Take for
example the problem of individuality in quantum mechanics (pp.
132–140; simplified to exclude considerations of field theories).
According to the authors, the fact that when in an entangled quan-
tum state individual particles can be individuated by neither their
intrinsic nor their extrinsic properties means that we are faced
with a choice between either individuating each particle with re-
spect to some distinct metaphysical ‘this-ness’, or abandoning
the notion of individuality altogether in favour of a structuralist
ontology. Unsurprisingly, the authors opt for the latter option,
and all but the most committed neo-scholastic would presumably
feel some sympathy here. Yet all this is a debate over the interpre-
tation of quantum mechanics; it is therefore hard to see how a nat-
uralistic injunction to reject any hypothesis that ‘conflicts with
fundamental physics’ could decide the matter either way, as the
naturalistic metaphysician would have us believe. If primitive
this-ness conflicted with fundamental physics, it would never have
been a potential interpretation of fundamental physics in the first
place.

Ladyman et al. go on to argue that a structuralist ontology also
provides a reconciliation between quantum mechanics and gen-
eral relativity (p. 182), which the more metaphysically spurious
option does not, on the grounds that metaphysical problems in
both disciplines are dissolved by moving to an ontology of struc-
ture (thus fulfilling the aim of unification encoded in the Principle
of Naturalistic Closure). But the reconciliation here is conceded to
be nothing more than analogical (p. 143)—both quantum particles
and space-time points are to be denied individuality in favour of
structure. Yet it remains mysterious how exactly such structures
are supposed to relate to one another. Pending such a develop-
ment, one feels that one could just as well argue that haecceities
provide a reconciliation between quantum mechanics and general
relativity, on the grounds that both quantum particles and space-
time points are to be individuated by a primitive this-ness.
Of course, a metaphysics of haecceities and their like may well
strike one as empirically surplus and ontologically extravagant;
but again, these are metaphysical desiderata, not something one
finds encoded in scientific methodology. David Lewis, our favour-
ite neo-scholastic stalking horse, can offer us good grounds here
(in terms of a good old fashioned cost-benefit analysis of ontolog-
ical commitment) for rejecting an ontology of haecceities in favour
of an ontology of structure; yet awkwardly, the naturalistic meta-
physician cannot.

To reiterate: Ladyman et al. provide a compelling case for ontic
structural realism, especially with respect to the philosophy of
physics, and an ingenious reconciliation of fundamental and spe-
cial science. Their invective against neo-scholastic metaphysics is
also well taken. My concern however is with the compatibility of
these two central themes: metaphysics without science may in-
deed be blind; yet without its guiding presumptions, institutional
constraints, theoretical entrenchment and methodological arbitra-
tion, one wonders whether science without metaphysics is also
empty.
ue causal theory of reference here; as they put it, ‘the debate about whether the world
nguage. No matter how the realist contrives a theory of reference so as to be able to say
l of space’ (p. 92).



P. Dicken / Stud. Hist. Phil. Sci. 39 (2008) 290–293 293
Acknowledgement

Many thanks to Tim Lewens and Peter Lipton, with whom I dis-
cussed various aspects of this paper; and to the Master and Fellows
of Churchill College, Cambridge, where I completed this work as a
Research Fellow.

References
Armstrong, D. (1983). What is a law of nature? Cambridge: Cambridge University

Press.
Bain, J., & Norton, J. D. (2001). What should philosophers of science learn from the

history of the electron. In J. Z. Buchwald & A. Warwick (Eds.), Histories of the
electron (pp. 451–465). Cambridge, MA: MIT Press.

Bourne, C. (2006). A future for presentism. Oxford: Oxford University Press.
Brown, H. (1993). Correspondence, invariance and heuristics in the emergence of

special relativity. In S. French, & H. Kaminga (Eds.), Correspondence, invariance
and heuristics: Essays in honour of Heinz Post (pp. 227–260). Boston Studies in
the Philosophy of Science, 148. Dordrecht: Kluwer.

Dennett, D. (1991). Real patterns. Journal of Philosophy, 88, 27–51.
Dupré, J. (1993). The disorder of things. Cambridge, MA: Harvard University Press.
French, S. (1989). Identity and individuality in classical and quantum physics.

Australasian Journal of Philosophy, 67, 432–446.
French, S. (1998). On the withering away of physical objects. In E. Castellani (Ed.),

Interpreting bodies: Classical and quantum objects in modern physics
(pp. 93–113). Princeton: Princeton University Press.

French, S., & Ladyman, J. (1997). Superconductivity and structures: Revisiting the
London account. Studies in History and Philosophy of Modern Physics, 28,
363–393.
French, S., & Ladyman, J. (2003). Remodelling structural realism: Quantum
physics and the metaphysics of structure. Synthese, 136, 31–56.

Kim, J. (1998). Mind in a physical world. Cambridge, MA: MIT Press.
Ladyman, J. (1998). What is structural realism? Studies in History and Philosophy

of Science, 29, 409–424.
Ladyman, J. (2000). What’s really wrong with constructive empiricism? Van

Fraassen and the metaphysics of modality. The British Journal for the
Philosophy of Science, 51, 837–856.

Lewis, D. (1999). Humean supervenience debugged. In idem, Papers in
metaphysics and epistemology (pp. 224–247). Cambridge: Cambridge
University Press.

Lyre, H. (2004). Holism and structuralism in U(1) gauge theory. Studies in History
and Philosophy of Modern Physics, 35, 643–670.

Monton, B., & van Fraassen, B. C. (2003). Constructive empiricism and
modal nominalism. The British Journal for the Philosophy of Science, 54,
405–422.

Putnam, H. (1967). Time and physical geometry. Journal of Philosophy, 64,
240–247.

Ross, D. (2000). Rainforest realism: A Dennettian theory of existence. In D. Ross,
A. Brook, & D. Thompson (Eds.), Dennett’s philosophy: A comprehensive
assessment (pp. 147–168). Cambridge, MA: MIT Press.

Saunders, S. (1993). To what physics corresponds. In S. French, & H. Kamminga
(Eds.), Correspondence, invariance and heuristics: Essays in honour of Heinz Post
(pp. 295–325). Boston Studies in the Philosophy of Science, 148. Dordrecht:
Kluwer.

Stanford, K. (2003). Pyrrhic victories for scientific realism. Journal of Philosophy,
100, 551–572.

van Inwagen, P. (1990). Material beings. Ithaca, NY: Cornell University Press.
Worrall, J. (1989). Structural realism: The best of both worlds? Dialectica, 43,

99–124.


	Conditions may apply
	Acknowledgement
	References