© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

METAPHILOSOPHY
Vol. 50, No. 3, April 2019
0026-1068

A NEW TASK FOR PHILOSOPHY OF SCIENCE

NICHOLAS MAXWELL

Abstract: This paper argues that we philosophers of science have before us an

important new task that we urgently need to take up. It is to convince the scien-
tific community to adopt and implement a new philosophy of science that does
better justice to the deeply problematic basic intellectual aims of science than
that which we have at present. Problematic aims evolve with evolving knowl-
edge, that part of philosophy of science concerned with aims and methods thus
becoming an integral part of science itself. The outcome of putting this new
philosophy into scientific practice would be a new kind of science, both more
intellectually rigorous and one that does better justice to the best interests of
humanity.

Keywords: philosophy of science, physics, revolution for science, revolution for

philosophy of science, metaphysics of science, problematic aims of science,
natural philosophy, standard empiricism, aim-oriented empiricism, physicalism,
scientific progress.

1. Introduction
We philosophers of science have before us an important new task that
we urgently need to take up. It is to convince the scientific community to
adopt and implement a new philosophy of science that does better justice
to the deeply problematic basic intellectual aims of science than that
which we have at present. The outcome of putting this new philosophy
into scientific practice would be a new kind of science, both more intel-
lectually rigorous and one that does better justice to the best interests
of humanity. It will be, I admit, a hard task to convince the scientific
community that the conception of science they tend to take for granted
needs to be radically improved. Not all scientists hold the philosophy
of science in high esteem. We need, nevertheless, to do what we can to
get across the argument that a new conception of science needs to be
adopted and implemented, one that acknowledges and seeks to improve
problematic aims of science.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 317

2. The Problematic Aims of Science

The crucial question that has to be decided to determine the extent to
which the philosophy of science has fruitful implications for science it-
self is simply this: Are the basic aims of science unproblematic and fixed?
Or are they problematic, permanently in need of improvement, so that
the aims and associated methods of science need to be improved as scien-
tific knowledge itself improves? If the former holds, there is no reason
why the activity of articulating the aims and methods of science—the
philosophy of science, in other words—should have much of an impact
on science itself.1 The task is simply to get clear what the basic aim of
science is, and what the basic methods are, once and for all. This task
may well be what might be called a “meta” task, one that leaves science
itself unaffected. The task is to improve our knowledge and understand-
ing of the nature of the scientific enterprise, but not to change science
itself. That the philosophy of science, conceived in these terms, has noth-
ing much to offer science itself is no cause for shame.
But the situation is radically changed if the aims of science are inher-
ently problematic, permanently in need of improvement as scientific
knowledge and understanding improve, so that there is something like
positive feedback between improving scientific knowledge and improv-
ing aims and associated methods—improving knowledge about how to
improve knowledge. Philosophy of science—construed as the enterprise
of articulating and justifying the aims and methods of science—becomes
a vital, integral part of science itself. As science evolves, it influences the
philosophy of science; and vice versa, as philosophy of science evolves, it
influences science. Or at least that is what ought to go on.
Let us give a bit more flesh to these very abstract considerations. I now
formulate two rival philosophies of science that give starkly opposing
answers to the above question. The first holds that science has a fixed
aim, and fixed associated methods; the second that science has profoundly
problematic, and so evolving, aims, and thus evolving associated meth-
ods. The first implies that philosophy of science is a meta-discipline, one
that leaves science essentially unaffected—just as astronomical study of
the moon leaves the moon unaffected. The second implies that science and
philosophy of science interact with each other in both directions.
But first, a few words about the aims of science. Our concern here is very
definitely with the aims of science , not the aims of scientists. Furthermore,
our concern is not with what the scientific community declares to be the
aims of science; rather, what matters is aims that are implicit in the actions
of the scientific community—aims that, it seems, scientific actions strive
to attain. Questions about the aims of science arise in two contexts: the
1
Philosophy of science does much more than seek to articulate the aims and methods of
science, and justify them. In what follows I am primarily concerned with that part of philos-
ophy of science concerned with the aims and methods of science.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

318 NICHOLAS MAXWELL

context of discovery and the context of justification—the context of

determining what is, and what is not, a contribution to science. Our con-
cern here is primarily with the aims of science in the latter context.
How are rival proposals as to what constitutes an aim for science to be
assessed? What kinds of consideration can be brought to bear in deciding
whether such and such a proposed aim is acceptable? I suggest that there
are at least the following three considerations.

1. The proposed aim must be such that it does reasonable justice

to relevant actual scientific actions—acceptance and rejection of
potential contributions to science. It must be possible, in other
words, to interpret these actions, in a reasonably good way, as
attempts to realize the aim in question—so that attributing the
aim to science makes reasonably good sense of what scientists
do in practice. (We do not require that the aim make sense of
everything scientists do, in every detail and respect.)
2. The proposed aim must be sufficiently worthwhile. It must do justice
to the intellectual value of science.
3. It must be possible to show that scientific actions—contributions
that are accepted—do indeed constitute steps towards realizing the
aim, or at least that there are sufficiently good reasons to hold that
this is the case. The aim is, in other words, realizable, as far as we
know, and contributions to science do indeed constitute steps to-
wards realizing the aim, or at least we have reasonably good reasons
to hold that this is the case.

The third of these three requirements is the most controversial. In order

to meet it, we need to solve the problem of induction. I will have more to
say about this issue below.
Here, now, are my two rival philosophies of science. They are, more
accurately, two rival philosophies of physics; or perhaps even more accu-
rately, two rival philosophies of theoretical physics.
Standard empiricism. The basic aim of physics is truth, nothing being
presupposed about the truth. The basic method is to assess claims to
knowledge impartially with respect to evidence. When it comes to deciding
what laws and theories to accept, considerations of simplicity, unity, or
explanatory power may legitimately influence choice of law or theory in
addition to empirical success and failure, but not in such a way that nature
herself, or the phenomena, are presupposed to be simple, unified, or com-
prehensible. Again, choice of theory may, for a time, be biased in the direc-
tion of some metaphysical view, Kuhnian paradigm, or Lakatosian “hard

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 319

core.” The decisive point is that no permanent thesis about the world can be
accepted as a part of scientific knowledge independent of evidence .2
Aim-oriented empiricism. The basic aim of physics is to discover the
truth presupposed to be physically explanatory or comprehensible . That is,
the basic aim of physics presupposes that an underlying, unified, physical
something exists in nature, inherent in all phenomena, that suffices to
explain everything that occurs. This aim is profoundly problematic because
it is a sheer metaphysical conjecture that the universe is comprehensible in
the sense indicated. The basic method comes in two parts. Part 1 subjects
the profoundly problematic metaphysical conjecture about the physical
comprehensibility of the universe, inherent in the basic aim of physics, to
sustained critical scrutiny, rival possible versions of the conjecture being
articulated and assessed, that version of the conjecture being provisionally
accepted that leads, or promises to lead, to the most empirically progres-
sive research programme. Part 2 involves accepting those theories that are
sufficiently in accord with (a) evidence and (b) the best available metaphys-
ical conjecture concerning the physical comprehensibility of the
universe.3
Let us put to one side, for the moment, objections to these two views,
questions about which is to be preferred, or disliked least. The key point I
want to emphasize is that these two rival philosophies of physics have very
different implications for how physics and the philosophy of physics ought
to be inter-related.
Given standard empiricism, with its fixed aim for physics, and with its
broadly fixed methods, there is not much room for that fragment of phi-
losophy of physics concerned with the aims and methods of physics to
have much of an impact on physics itself. The task of the philosophy of
physics is to make explicit what is presumably implicit in scientific practice:
make explicit the basic aim, the methods adopted in pursuit of that aim,
and then provide a justification for the methods in question. The latter
involves at least demonstrating that the methods in question are the best to
adopt granted that one seeks to realize the specified aim. None of this has
implications for how physics itself should be conducted—not unless phys-
icists fail in practice to put the specified methods into practice. Physics and
the philosophy of physics, granted standard empiricism, seem sharply dis-
tinct. This is indeed declared to be the case by standard empiricism itself.

2
This rather meagre thesis of standard empiricism is common ground for such otherwise
diverse doctrines as logical positivism, inductivism, logical empiricism, hypothetico-deduc-
tivism, conventionalism, constructive empiricism, pragmatism, realism, inference-
to-the-best-explanationism, the views of Popper, Kuhn, and Lakatos, and many more recent
views as well: see Maxwell 1998, 38–45; 2004, chap. 1, nn. 5, 6, and 14). For discussion of the
claim that Popper, Kuhn, and Lakatos defend versions of standard empiricism, see Maxwell
2005.
3
For expositions of aim-oriented empiricism, see Maxwell 1974; 1998; 2004; 2005; 2007,
chap. 14; 2013; 2017a; 2017b; 2019b; 2017c, chap. 3.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

320 NICHOLAS MAXWELL

The basic method of standard empiricism is to assess claims to knowledge

by means of evidence. Anything that is not open to being so assessed can-
not be a part of science. But the philosophy of science—or, more spe-
cifically, the philosophy of physics—is a view about what the basic aim
and methods of physics ought to be. In being concerned with norms or
ideals, it is not the sort of entity that is empirically testable. Thus, granted
standard empiricism, the philosophy of physics is not a part of physics
itself. This conclusion is borne out, in particular, by Karl Popper’s demar-
cation requirement, which holds that a discipline, in order to be scientific,
must consist of empirically falsifiable theories or statements (Popper 1959,
40–42). The philosophy of physics does not consist of such theories; hence
it is not itself a part of physics. Standard empiricism itself drives a wedge
between science and the philosophy of science.
All this changes dramatically if we move from standard to aim-oriented
empiricism. According to aim-oriented empiricism, there are profoundly
problematic metaphysical theses about the nature of the universe inher-
ent in the aims of physics. These presupposed theses, in the more or less
specific form that they are accepted at any stage in the development of
physics, are almost bound to be false. And yet—according to aim-oriented
empiricism—they exercise a profound influence over the kind of theory
that is sought in the context of discovery, and the kind of theory that is
accepted in the context of justification. If physics is to meet with success,
it needs to make good choices of basic metaphysical assumptions, and yet
it is here that physics is almost bound to get things very seriously wrong,
since we are concerned with mere unfounded conjectures —wild specula-
tions —about that of which we are most ignorant, the ultimate nature of
the universe. And if we glance at the past, we see that physics has indeed
got things very wrong, again and again, in that it has blundered from one
grossly false metaphysical speculation to another. In the seventeenth cen-
tury there was the idea that the universe is made up of corpuscles that
interact by contact. This gave way to the Boscovichean idea that it is made
up of point particles with mass, surrounded by rigid spherically symmet-
rical forces that act at a distance; this in turn gave way to the idea that it
is made up of charged point particles embedded in a classical field, which
in turn became the idea that there is just one self-interacting, unified field.
This became the idea that the universe is compounded of quantum entities
(whatever they may be), which in turn became a unified quantum field,
and then quantum strings in ten or eleven dimensions of space-time. To
repeat: if physics is to meet with success, it must make a good choice of
basic metaphysical assumption, but it is just here that it is bound to get
things drastically wrong. The only hope—since we cannot avoid making
some assumption—is to subject such assumptions to sustained critical
scrutiny and development, in an attempt to improve the specific assump-
tion that is adopted.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 321

This crucial activity of imaginatively developing and critically assess-

ing metaphysical assumptions for physics in an attempt to improve the
specific assumption that is adopted is thus likely to have a profound influ-
ence over theoretical physics itself, in the contexts of both discovery and
justification, whether for good or for ill. If physics makes a good choice of
metaphysical conjecture, it may well forge ahead; if it makes a bad choice,
it will be stultified. But this enterprise of exploring possible metaphysical
assumptions for physics is the philosophy of physics. It is concerned with
empirically untestable, metaphysical ideas. It seeks to improve metaphys-
ical conjectures implicit in the basic aim of physics, thus improving the
permanently problematic aim. And in seeking to improve metaphysical
assumptions, it seeks also to improve associated methods .
Thus, according to aim-oriented empiricism, the philosophy of phys-
ics—or that central part of the philosophy of physics concerned with the
aims and methods of physics—is (or ought to be) a vital, integral part of
physics itself, exercising a very substantial influence over physics in the
contexts of both discovery and justification.4

3. A Fundamental Watershed for Science and Philosophy of Science

The key difference between standard empiricism and aim-oriented em-
piricism has to do with whether the basic aims of science do not, or do,
have problematic assumptions inherent in them. Standard empiricism
declares that the basic intellectual aim is factual truth. How we get it,
how we can know whether or not we have got it: these questions are pro-
foundly problematic. But the aim itself is not problematic. It is fixed:
factual truth in so far as it can be attained.5 “Snow is white” is true if and
only if snow is white: that is all there is to it. There are no substantial
assumptions inherent in the aim all too likely to be false that need per-
sistent revision and improvement.
Aim-oriented empiricism, by contrast, holds that there is a substantial
and profoundly problematic metaphysical assumption built into the basic
aim of physics: the universe is physically comprehensible. It is such that there
is a true physical “theory of everything” that is unified. This assumption,
and more specific versions of this assumption, may well be false. As we have
seen, the big, problematic assumption inherent in the aim of physics may
well need to be modified and improved, again and again, as physics proceeds.
This key dividing line between standard and aim-oriented empiricism
constitutes a quite fundamental watershed for science, for philosophy of

4
For a more detailed exposition and defence of these points, see Maxwell 2017a and
2017b.
5
There is here room for slight disagreement. Scientific realists hold that science can ac-
quire knowledge of unobservable entities, whereas instrumentalists hold that scientific
knowledge is restricted to observable phenomena.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

322 NICHOLAS MAXWELL

science, and for the relationship between the two. Stand anywhere on the
standard empiricist side of the watershed (upholding any one of the well-
known views about science) and one will hold that science has a fixed basic
intellectual aim and fixed methods.6 Science and philosophy of science are
distinct. Philosophy of science is a meta-discipline, studying science. It is
not the job of philosophers of science to tell scientists how to do science.
Stand anywhere on the aim-oriented empiricist side of the watershed, and
all this changes. Philosophy of science needs persistently to develop and
critically assess new possible metaphysical assumptions, inherent in new
possible aims for physics, in an attempt to improve the assumption that is
accepted, the aim that is pursued. In doing this, philosophy of physics
thereby seeks to improve associated methods of physics (methods employed
in deciding what theories are to be accepted and rejected, along with
empirical methods). As we have seen, this activity of attempting to improve
problematic aims and methods of physics, partly in the light of improving
scientific knowledge and understanding, is almost bound to have major
implications for physics itself. Indeed, philosophy of physics construed in
this way—the enterprise of attempting to improve problematic aims and
methods of physics—is a vital, integral part of physics itself, influencing
and being influenced by what goes on in the rest of physics, both experi-
mental and theoretical.
These dramatic differences in the whole character of the philosophy of
physics and physics itself, and the relationship between the two, all devolve
from this key issue: Are there, or are there not, problematic assumptions
inherent in the basic aim of physics? Does the basic aim need to evolve, to
be improved, or is it unproblematic and fixed?

4. Basic Argument
I can now state, in a little more detail, the first step of the central argu-
ment of this paper. Standard empiricism as characterized above is taken
for granted, in one or other version, by most scientists and philosophers
of science.7 It is, however, untenable. Physics in particular and natural
science more generally cannot proceed in accordance with its edicts.
Given any accepted physical theory, T, endlessly many rivals can be con-
cocted to fit the facts even better than T. Thus, granted that T is
Newtonian theory, one rival theory might be Newtonian theory modi-
fied so that (a) it has an additional, independently testable and corrobo-
rated postulate that successfully predicts phenomena that T cannot
predict, and (b) predicts that gravitation will become a repulsive force in

6
This holds for such versions of standard empiricism as inductivism, positivism, conven-
tionalism, hypothetico-deductivism, Bayesianism, constructive empiricism, the views of
Popper, Kuhn, Lakatos, and others.
7
See note 2 and Maxwell 2017a, 73–74.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 323

the first moment of 2050, so that F = – Gm1m2/d2. Another rival might (a)
have the additional empirically successful postulate and (b) predict that
for gold spheres of mass greater than a thousand tons in outer space, the
law of gravitation has the form F = Gm1m2/d3. Endlessly many empiri-
cally more successful rivals to Newtonian theory can be concocted along
these lines. In practice, these are all ignored because they are hopelessly
ad hoc, complex, disunified, or non-explanatory. But in only ever accept-
ing unified theories, and ignoring endlessly many empirically more suc-
cessful disunified rivals (of the kind just considered), physics thereby
makes a substantial, implicit assumption about the nature of the uni-
verse: it is such that all seriously disunified theories are false. The uni-
verse is such that empirically more successful disunified rivals to T will
make false predictions where they clash with the predictions of T. Not
only does physics make a substantial metaphysical assumption about the
universe in persistently accepting only unified theories and ignoring end-
lessly many empirically more successful disunified rivals; it must make
some such assumption. If no assumption is made, and theories are as-
sessed impartially with respect to evidence, physics would be over-
whelmed by an infinite swamp of empirically successful but horribly
disunified theories. Thus, standard empiricism cannot be implemented
in scientific practice. If some kind of requirement of simplicity, unity, or
explanatoriness is invoked in addition to empirical considerations when
it comes to deciding what theories are to be accepted, that means some
substantial metaphysical assumption is made about the universe—and
that contradicts standard empiricism. Standard empiricism only be-
comes scientifically viable if it is contradicted in practice!8
How is it possible for physics to have been so astonishingly successful if
most physicists take for granted the untenable doctrine of standard empir-
icism? The answer is that they have not taken it too seriously in scientific
practice. In practice, considerations that have to do with the simplicity,
unity, explanatory, or non-ad-hoc character of a theory are taken very
seriously indeed in deciding what is to be accepted and rejected, in addi-
tion to empirical considerations. Theoretical physicists seek to develop
conservation, invariance, and symmetry principles capable of determining
what theories are to be accepted, along with empirical considerations.9 All

8
For more detailed expositions of this refutation of standard empiricism, see Maxwell
1998, chap. 2; 2017a, 69–83; 2017b.
9
Lorentz invariance, gauge invariance, local gauge invariance, charge, parity and time
reversal symmetry, and supersymmetry are all symmetry principles that have played an im-
portant role in recent times in theoretical physics. These symmetry principles are, in effect,
methodological rules: they specify what features a theory must have if it is to be a candidate
for acceptance. At the same time, they are methodological principles that have a fallible,
conjectural aspect to them: they may need to be rejected (and some have been rejected). This
accords with aim-oriented empiricism but is starkly at odds with standard empiricism (which
does not really have a role for evolving methodological rules).

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

324 NICHOLAS MAXWELL

these non-empirical requirements a theory must satisfy to be accepted in

effect appeal to implicit metaphysical principles.10 Standard empiricism is
quietly and radically violated in scientific practice, without this being
acknowledged. It is this that has made possible progress in physics—and
progress in science more generally.11 As Einstein once declared, “If you
want to find out anything from the theoretical physicist about the methods
they use, I advise you to stick closely to one principle: don’t listen to their
words, fix your attention on their deeds” (1973, 270).
Philosophers of science, however, have taken standard empiricism
much more seriously. The whole way the discipline is conceived and pur-
sued, the way its relationship to science is conceived and maintained,
presupposes standard empiricism. All the diverse, well-known doctrines
about science that philosophers of science have developed are versions of
standard empiricism, as I have already indicated. It is this that accounts
for the scientific poverty of academic philosophy of science, its irrelevance
to science, so cruelly pointed out by scientists themselves. The intellectual
activity of developing, exploring, and critically assessing possible aims for
physics, possible metaphysical assumptions inherent in these aims—an
activity that constitutes such a vital, integral part of physics if one stands
on the aim-oriented empiricist side of the watershed—does not make sense,
and is not a viable or coherent endeavour for philosophy of physics if one
stands on the standard empiricist side of the watershed. On this side of the
watershed, the basic aim of physics is fixed; it does not make sense to try
to improve it. Physics does not make problematic metaphysical assump-
tions concerning the physical comprehensibility of the universe, at least
not in the context of justification, and so there is no philosophical task of
trying to improve these assumptions. Such assumptions may be made in
the context of discovery, but that, viewed from this side of the watershed,
is physics , not philosophy of physics . Standard empiricism thus cripples
the philosophy of science and renders it more or less irrelevant from the
standpoint of science itself—as scientists themselves have attested.
Put in Kuhnian terms, philosophy of science has long been in a state of
crisis, at least since the heady days of logical positivism. A revolution is
long overdue. This would involve a radical change of paradigm, from

10
In demanding that a theory satisfy such a requirement to be accepted, physics implicitly
accepts the thesis that the universe is (or the phenomena are) such that a theory, in order to
be near enough to the truth to be acceptable, must satisfy the requirement. In other words,
the universe is (or the phenomena are) such that all theories that fail to satisfy the require-
ment are false.
11
As a result of taking standard empiricism as seriously as they do, however, physicists
cannot take up, in the explicit, sustained way that is required, the articulation and critical
assessment of metaphysical conjectures implicit in the symmetry principles they propose,
consider, and adopt. For to do so would be to violate the precepts of standard empiricism all
too blatantly. For an informal and entertaining account of the confusion that results, see
Hossenfelder 2018. See also Maxwell 2017a, chap. 5.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 325

standard to aim-oriented empiricism. The outcome would be a transfor-

mation in the nature of the discipline. Both science and philosophy of
science would change, and the relationship between the two would change
as well. As I have put it elsewhere, science and philosophy would be
brought together to create a modern version of natural philosophy.12

5. Recent Work on the Metaphysics of Science

In recent years the metaphysics of science has become a hot topic in the
philosophy of science. Does this work acknowledge that metaphysics
plays a role in science only in the context of discovery, a metaphysical
thesis being acceptable only if it is compatible with current scientific
knowledge, there being no hint of “the basic argument” of section 4
for aim-oriented empiricism, some version of standard empiricism thus
being presupposed? Or is some version of “the basic argument” pro-
pounded, aim-oriented empiricism thus being defended, there being a
full recognition of the point that the aims and methods, and thus the
metaphysics too, of science evolve with evolving scientific knowledge?
In other words, on which side of the watershed between standard and
aim-oriented empiricism does this body of work stand? Let us see.
Dilworth 2007 is, in some respects, a pioneer as far as this relatively
recent body of work is concerned—the book was first published in 1996.
There is a clear recognition that metaphysical principles play an important
role in science. However, despite several references to Maxwell 1984, where
aim-oriented empiricism is expounded and defended, there is no hint in
Dilworth’s book of “the basic argument” for, and key tenets of, aim-ori-
ented empiricism. For a critical appraisal see Maxwell 2009.
Ladyman et al. 2007 decisively criticises that enterprise of analytic phi-
losophy that seeks to do metaphysics independently, or in ignorance, of
modern physics. It is clearly recognized that the task of scientific meta-
physics is to provide a basis for the unification of two or more accepted
fundamental theories of physics, all such proposals being conjectural and
likely to be false. The account of unification is, however, unsatisfactory—
it takes Kitcher 1976 and 1981 for granted.13 There is no hint of the con-
ception of unification that is required: see Maxwell 1998, chaps. 3–4, and
2004, 160–74. Much more seriously, there is no hint of “the basic argu-
ment” of section 4 above for aim-oriented empiricism. Far from explicat-
ing something like the hierarchical meta-methodology of aim-oriented
empiricism (see below), designed to subject the metaphysics of physics to
sustained critical scrutiny and attempted improvement, Ladyman et al.
2007 actually states at one point that “there is no such thing as ‘scientific

12
See Maxwell 2017a. See also Maxwell 1984, chap. 9; 1998; 2004.
13
For a decisive criticism of Kitcher 1976 and 1981, see Maxwell 1998, 62–68.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

326 NICHOLAS MAXWELL

method’” (27). Like Dilworth, Ladyman and colleagues do not take up

“the new task for philosophy of science” and do not argue that it should
be done.
Chakravartty (2007) is concerned to articulate a version of scientific
realism that does justice both to modern science and to the critics of real-
ism, but all within the framework of standard empiricism, which is taken
for granted throughout. There is not a hint of “the basic argument,” nor
of the need for a new hierarchical meta-methodology for physics designed
to facilitate the improvement of the metaphysics of physics as science
proceeds.
Lange (2009) is concerned to clarify what it is in existence that renders
a law of nature, even though contingent, nevertheless in some sense “nec-
essary,” and distinct from a true accidental generalization.14 Standard
empiricism is presupposed without discussion.
Morganti (2013) seeks to develop a metaphysics for science that steers a
middle course between scientific naturalism and philosophical a priorism.
There is no hint of aim-oriented empiricism or the new task for philoso-
phy of science.
In their excellent introduction, Mumford and Tugby begin by stating
that “science can only exist in an ordered, patterned world, and it is argued
that the core aim of the metaphysics of science is to investigate the nature
of that order” (2013, 3). This promising beginning does not, however, lead
on to the point that this metaphysical thesis of patterned order is a prob-
lematic presupposition of physics, very likely to be false in the specific
form adopted at any given stage in the development of physics, a new
hierarchical methodology for physics being required to subject the thesis
to sustained criticism and attempted improvement.
Kincaid, Ladyman, and Ross 2013 is a collection of essays devoted to
the idea that metaphysics must be based on modern science. The book
contains contributions from Harold Kincaid, Anjan Chakravartty, Paul
Humphreys, Andrew Melnyk, Daniel Dennett, James Ladyman and Don
Ross, Mark Wilson, Michael Friedman, and Jenann Ismael. As Kincaid
mentions in the introduction, many of these contributors have recently
published books on scientific metaphysics. It is thus highly significant that
nowhere do we find, in this collection, any mention of “the basic argu-
ment” that leads one to acknowledge that physics makes a highly problem-
atic metaphysical presupposition concerning underlying unity in nature, a
new meta-methodology being required to subject this presupposition to
sustained scrutiny and attempted improvement. The first steps towards
“the new task for philosophy of science” are nowhere taken.15

14
For another view on this issue, see Maxwell 1968. See also Maxwell 2019a, chap. 1.
15
For a much more detailed critical appraisal of the above-mentioned works of all these
authors, apart from Lange 2009, along the lines indicated here, see Maxwell 2019a, chap. 4.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 327

What all these authors ignore is a central source for the metaphysics of
physics that comes, not from physical theory, but from the methods of
physics—specifically that methodological rule that asserts: in order to be
acceptable, a fundamental physical theory must be (sufficiently) unified .16
It is the persistent acceptance of unified theories only, when endlessly
many empirically more successful disunified rivals are available, that com-
mits physics to the metaphysical presupposition that, at the very least, the
universe is such that all disunified theories are false. Recognition of this
point constitutes the key step towards adopting the hierarchical method-
ology of aim-oriented empiricism (see below), and the new task for philos-
ophy of science. I have developed this argument in detail in a series of
publications since 1974—see Maxwell 1974, 1984, 1998, 2004, 2005, and
elsewhere17—and yet none of the above publications refers to this body of
work, apart from Dilworth 2007.18 On the other hand, those who have
read these publications of mine speak well of them: see, for example,
Kneller (1978, 80–87 and 90–91); Longuet-Higgins (1984); Smart (2000);
Cory (2000); McHenry (2000); Roush (2001); Muller (2004); and
MacIntyre (2009).

6. An Improved Version of Aim-Oriented Empiricism

Aim-oriented empiricism has one crucial advantage over standard empir-
icism. Whereas the latter denies, falsely, that physics makes a substantial
metaphysical assumption about the nature of the universe, aim-oriented
empiricism, correctly, acknowledges that such an assumption is made,
recognizes that it is a mere conjecture all too likely to be false in the
specific version accepted at any stage in the development of physics, and
throws it open to sustained critical scrutiny in an attempt to improve it.
Aim-oriented empiricism, as formulated briefly above, still faces prob-
lems, however. What possible justification can there be for physics to
presuppose that the universe is physically comprehensible in the sense
that the true physical theory of everything is unified ? What does it mean,
in any case, to say of a theory that it is unified? How can metaphysical
theses about the comprehensibility of the universe be improved ? What if

16
Mumford and Tugby (2013, introduction) come closest to recognizing this crucial
point.
17
See especially Maxwell 2017b; also 2017a and 2017c.
18
Additional recent works on the metaphysics of science I have examined that take stan-
dard empiricism for granted and fail even to mention, let alone discuss, aim-oriented empir-
icism dating back to Maxwell 1974 include: Ellis 2001; Lowe 2006; Bird 2007; M. O’Rourke,
M. H. Slater, A. Borhini, P. Godfrey-Smith, N. Latham, R. Sorensen, A. C. Varzi, M. Devitt,
B. Nany, N. E. Williams, B. Glymour, N. G. Rheins, J. K. Crane, R. Sandler, and K. Vihvelin
in Campbell et al. 2011; Trout 2016; S. Yudell, K. Brading, M. Strevens, C. K. Waters, K.
Stanford, J. Saatsi, and M. Thomas-Jones in Slater and Yudell 2017.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

328 NICHOLAS MAXWELL

the universe is not physically comprehensible? Does that just mean that
physics becomes impossible?
In order to solve these problems, the version of aim-oriented empiricism
briefly formulated above needs to be radically developed and improved.
Elsewhere, I have developed and improved the view in some detail along
these lines, and I have shown in detail how it solves these (and related)
problems, so here I will be brief.
Aim-oriented empiricism needs to be formulated in such a way that
physics makes, not just one metaphysical assumption, but a hierarchy of
such assumptions: see figure 1. As one goes up the hierarchy, assumptions
become increasingly insubstantial, and so increasingly likely to be true,

FIGURE 1. Aim-oriented empiricism

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 329

and increasingly such that their truth is required for science, or the pursuit
of knowledge, to be possible at all.
At the top there is the thesis that the universe is such that we can con-
tinue to acquire knowledge of our immediate environment sufficient to
make life possible. Next down, at level 6, there is the thesis that the uni-
verse is such that we can formulate a conjecture about it that enables us to
improve our methods for improving knowledge. These two assumptions
are accepted permanently, not because we have any reason to hold them
to be true, but because, granted we seek to improve knowledge of truth,
it can only help and cannot hinder the search for knowledge. Next down,
at level 5, there is the thesis that the universe is comprehensible in some
way or other. There is some one kind of explanation for all phenomena.
Everything that occurs does so because a society of gods wills it, or one
God wills it, or because it goes towards the realization of some cosmic
goal, or because it is in accordance with some cosmic computer pro-
gramme, or because it occurs in accordance with some unified pattern of
physical law—or because some other entity is, in some way, responsible for
everything that occurs. This has the merit of exemplifying the thesis one
up in the hierarchy: if the thesis is true, then we may progressively improve
our methods for the improvement of knowledge (at levels 1 and 2), by
honing in on that version of the thesis that best simulates the growth of
empirical knowledge. Next down, at level 4, there is the thesis that every-
thing occurs in accordance with a unified pattern of physical law. This
version of the thesis that the universe is comprehensible—physicalism , as
it may be called—has been astonishingly fruitful in stimulating the growth
of knowledge in physics, at levels 1 and 2. All the great developments in
theoretical physics since Kepler and Newton exemplify theoretical unifi-
cation , and thus this level-4 thesis. Newtonian theory unifies Kepler and
Galileo, terrestrial and astronomical phenomena. Maxwellian electrody-
namics unifies the electric and magnetic fields, and unifies light, infra-red,
radio, and ultra-violet rays, X-rays, and gamma rays in revealing they
are all electromagnetic waves of different wavelengths. Special relativity
brings greater unity to classical electrodynamics, partially unifies energy
and mass via the equation E = mc2, and partially unifies space and time by
means of Minkowski space-time. General relativity unifies space-time and
gravitation. Quantum theory and the atomic theory of matter unify a vast
array of laws concerning the physical and chemical properties of matter,
and the way matter interacts with light. Quantum electrodynamics unifies
quantum theory, classical electrodynamics, and special relativity; quan-
tum electroweak theory unifies quantum electrodynamics and the weak
force; quantum chromodynamics brings unity to the physics of hadrons.
All these theoretical developments enormously enhance the predictive
scope of theory and at the same time bring ever greater unity to physical
theory, thus drawing ever closer to capturing the level-4 metaphysical the-
sis in the form of a testable, unified physical theory of everything. In other

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

330 NICHOLAS MAXWELL

words, this level-4 thesis has been astonishingly empirically fruitful in sup-
porting an immensely empirically progressive research programme—the
whole enterprise of theoretical physics since Newton. At level 3 there is
the best available specific version of physicalism; today, this may be held
to be string theory. At level 2, there are the accepted fundamental theories
of physics, general relativity and the standard model—the quantum field
theory of fundamental particles and the forces between them. And at level
1 there is empirical phenomena, the low-level empirical laws of experimen-
tal results.
This hierarchical version of aim-oriented empiricism provides a frame-
work of relatively unproblematic assumptions and associated methods—
aims and methods—at levels 7 and 6, accepted and adopted permanently,
within which increasingly problematic aims and methods, as we go from
level 5 to level 3, may be critically assessed, developed, and, we may hope,
improved. At levels 5 to 3, that thesis is accepted which (a) best accords
with the accepted thesis above it in the hierarchy and (b) supports, or
promises to support, the most empirically progressive research programme
at levels 1 and 2. Theses at levels 7 and 6 are, we may conjecture, true; as
we descend from level 6 to level 3, we move from truth to falsity. The hier-
archy concentrates criticism and attempts at improvement where they are
likely to be most fruitful, low down in the hierarchy, and at the same time
provides a fixed framework, at levels 7 and 6, that restricts ways in which
theses, lower down in the hierarchy, can be modified to those ways likely to
be most fruitful from the standpoint of progress in physics.
This fixed framework facilitates something like positive feedback
between improving scientific knowledge and improving knowledge about
how to improve knowledge—improving assumptions and associated
methods, in other words. As we improve our knowledge of nature, we cor-
respondingly improve the nature of science. Everyone would accept that
this goes on at the experimental level: new knowledge leads to new exper-
imental techniques, new instruments such as the microscope, the telescope,
the particle accelerator. Aim-oriented empiricism specifies how this can go
on at the theoretical level, new theories leading to new metaphysical ideas,
and new metaphysical ideas leading to the discovery of new theories.19
This has occurred in the history of physics—or we would still be stuck
with Aristotle—but it would proceed in a much more explicit fashion, in a
way likely to be much more fruitful, if aim-oriented empiricism were
explicitly to be put into scientific practice.20
This hierarchical version of aim-oriented empiricism depicts the fusion
of physics and the philosophy of physics. It is vital, for the intellectual

19
For a detailed account of the fallible but rational method of discovery that aim-ori-
ented empiricism provides, and other fruitful implications for theoretical physics that the
view has, see Maxwell 2017a, chap. 5. See also Maxwell 1993.
20
This is spelt out in detail in Maxwell 2017a, chap. 5.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 331

rigour and success of the enterprise, that each should influence, and be an
integral part of, the other.21
I might add that the level-3 thesis and associated methods (level-3 aims
and methods) are methods for the discovery and acceptance of level-2 the-
ories;22 the level-4 aim and methods are meta-methods for the discovery
and acceptance of level-3 theses and associated methods; and so on as we
go up the hierarchy. Methods associated with the level-7 thesis are
meta-meta-meta-meta-methods! It is this meta-structure of the methods
of this hierarchical view that makes it possible for it to facilitate positive
feedback between improving knowledge and improving metaphysical the-
ses and associated methods—improving knowledge about how to improve
knowledge, in other words.
Elsewhere, I have shown that aim-oriented empiricism is both required,
and sufficient, to solve the problem of induction (see Maxwell 2017b, esp.
chap. 9). This is a decisive argument in support of the view, and against
all versions of standard empiricism. (Centuries of effort, ever since David
Hume, to solve the problem presupposing versions of standard empiri-
cism, have not met with success.)

7. Broader Implications
So far I have indicated a new task for the philosophy of physics: convince
physics that aim-oriented empiricism needs to be put into practice, and
then collaborate with theoretical physicists in doing just that. The scope
of the above argument can, however, be enlarged so that it becomes ap-
plicable, not just to physics, but to the whole of natural science.
In order for considerations analogous to the ones spelled out above in
connection with physics to apply to some other science, that science must
have permanently problematic aims , so that its aims need sustained critical
scrutiny in an attempt to improve them; furthermore, aims are improved in
such a way that associated methods need to be improved as well. If these
conditions obtain, then the activity of improving aims and methods—the
philosophy of science, in other words—needs to be pursued as an inte-
gral, influential part of the science itself. Wherever aims and methods need
to evolve with evolving science, a version of the hierarchical structure of
aim-oriented empiricism needs to be put into scientific practice. These
conditions hold in what follows. I have five points to make.
The first point to note is that physics is the fundamental natural sci-
ence, aspects of which are presupposed directly or indirectly by all other
branches of natural science. Thus chemistry, cosmology, astronomy, and

21
See Maxwell 2017a and 2017b. See also Maxwell 1984, chap. 9; 1998; 2004.
22
Elsewhere I have demonstrated that this hierarchical view provides physics with a ratio-
nal, if fallible, method for the discovery of new physical theories: see Maxwell 1993, 1998,
219–23; 2004, 191–205; and especially 2017a, chap. 5.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

332 NICHOLAS MAXWELL

geology all presuppose aspects of physics; biology and neuroscience pre-

suppose aspects of chemistry (and probably, in a direct but looser way,
aspects of physics as well). The above argument, ostensibly about physics
only, is actually about the whole of natural science.
Secondly, the above argument applies in a much more direct way to
all other branches of natural science. It is not just physics that makes
problematic assumptions inherent in its basic aim; this is true of all the
branches of natural science. Chemistry, molecular biology, genetics, neuro-
science, geology, astronomy, climate science: all make assumptions that are
composed partly (a) of items of knowledge from some more fundamental
science and partly (b) of conjectures about what there is for the science in
question to seek to discover and transform into scientific knowledge. The
latter component (b) is inherently and particularly problematic, in that it
concerns the domain of our ignorance, that which we conjecture, do not
yet know, but may come to transform into scientific knowledge. Genetics
seeks to discover what genes do and how they do it; cell biology seeks to
improve knowledge and understanding of what goes on in the cell; neuro-
science seeks to improve knowledge and understanding of how what goes
on in the brain is related to what happens to the person, what the person
does, thinks, perceives, remembers. And so on. No branch of science is
entirely ignorant of that which remains to be discovered; there are always
more or less inadequate conjectures or intimations about what exists to be
discovered, and what can be discovered by current methods. Thus, for each
branch of natural science, considerations that arise in connection with the-
oretical physics arise here too: each branch needs to make explicit prob-
lematic aims of the discipline, problematic assumptions inherent in these
aims, so that they may be critically assessed, developed, and, we may hope,
improved as the science proceeds. More or less specific, problematic aims,
and associated methods, of each scientific discipline need to be articulated,
critically assessed, developed, and improved as an integral part of the disci-
pline itself. Each discipline, in other words, needs to implement its own ver-
sion of aim-oriented empiricism—the philosophy of the discipline forming
an important, integral, influential part of the discipline itself. It may well
be an unnecessary extravagance for a science such as geology or astronomy
to put a version of aim-oriented empiricism into practice that has a hierar-
chical structure of seven levels of assumptions—like the seven levels of the
version of aim-oriented empiricism indicated above, applied to theoretical
physics. On the other hand, the two levels of standard empiricism, evidence
and theory, are definitely insufficient. Each scientific discipline needs to
acknowledge and represent at least three levels of sustained discussion:
evidence, theory, and aims —the latter including problematic assumptions
inherent in aims. Once a science is constituted and pursued in this fashion,
it brings together, and fuses , the science and its philosophy of science.
Thirdly, there is another, very important way in which the above consid-
erations apply to all the diverse branches of natural science, and to natural

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 333

science as a whole. The aims of science and the sciences are inherently
problematic, not just because they have problematic metaphysical and
other factual assumptions inherent in them; they are problematic because
they have value assumptions in them, and assumptions concerning social
use . The basic intellectual aim of science is not just truth , as standard
empiricism assumes; nor is it just explanatory truth (truth presupposed
to be inherently explanatory) as the version of aim-oriented empiricism
expounded above in sections 2 and 5 presupposes: more generally, it is
to improve knowledge of valuable truth , truth deemed to be significant,
useful, important, of interest or of value in some way, explanatory truth
being a special case of valuable truth. And valuable truth is sought so that,
once acquired, it can be used by people, either to enhance their knowledge
and understanding of the world around them for its own sake or to attain
other human goals by means of technological applications and in other
ways. There are, in other words, problematic humanitarian, social, even
political assumptions concerning the human use of science inherent in the
aims of science.
In order to become an item of scientific knowledge, it is not sufficient
that a result is new and very well established: it must reach a certain thresh-
old of significance, of value. One cannot contribute to science by counting
gravel on paths or leaves on trees—not unless this is a part of a broader,
more significant research programme. A science that accumulated a vast
store of well-established but irredeemably trivial facts would not thereby
be said to be making splendid progress. It is inevitable that values play
a role in deciding what does and what does not become a part of scien-
tific knowledge. The domain of fact is infinite; some aspects of the world
around us must be selected out as the significant aspects for science to con-
centrate on. And it is desirable that science should do this: we want science
to improve our knowledge about what matters or is of genuine interest or
use; we do not want knowledge of irredeemable trivia.
It is of course vital to appreciate that value judgements influence what
enters the domain of scientific knowledge in a way that is entirely different
from the way metaphysical theses concerning the comprehensibility and
knowability of the universe influence acceptance of physical theory, as
depicted above in sections 2 to 5. There, metaphysical assumptions influ-
ence judgements about truth . Values, however, ought not to influence
judgements about truth , since we have no reason to hold that just because
it would be good for something to be true, it is likely to be true, or just
because it would be bad, it is likely to be false.23 Values play a role in deter-

23
If values do influence judgements concerning epistemological merit, then this influence
should be negative : the more important it is for a result to be true, the more we desire it to be
true, the more there is at stake, then the harsher and more severe should be the experimental
examination of the result, the higher our standards for acceptance. This is especially the case
if lives are at stake, as in the use of new drugs.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

334 NICHOLAS MAXWELL

mining whether or not a given result should enter scientific knowledge—

whether it should be published in a scientific journal, for example in
influencing decisions about whether the result is sufficiently significant or
important to be published, to become a part of scientific knowledge.
Assumptions concerning value and social use are inherent in the aims
of science, and in the aims of diverse sciences, in contexts of both discov-
ery and justification. Here, the context of discovery is, perhaps, the more
important, for it is in this context that decisions are made about the prior-
ities of research, what research projects to fund and not to fund.
Value assumptions inherent in the aims of science, and in the aims of
specific sciences, are intrinsically and permanently problematic.24 Of value
to whom? In what kind of way? Of value in the immediate future, or in
fifty or a hundred years’ time? Who is to decide, and how? In addition to
the obvious, general problems that arise in deciding questions of value,
there are some very difficult questions that are specific to science. First, in
attempting to choose the best aims for scientific and technological
research, we seek those aims that are (a) possible to discover and (b) desir-
able or of value to discover. Both (a) and (b) are inherently problematic,
(b) especially. It is notoriously difficult to anticipate future uses and mis-
uses of scientific discoveries, ways in which discoveries will turn out to be
used for benefit and for harm.
Second, there is the permanent problem that science can be of value
in two very different ways: of value intellectually or culturally, in enhanc-
ing our knowledge and understanding of aspects of the world around us,
and of value technologically or practically, in enabling us to achieve other
things of value: food, health, shelter, transport, communications, and all
the benefits of the modern world. How are these two very different kinds
of value to be weighed up against each other? How is the value of pure sci-
ence to be weighed against the value of saving lives and alleviating human
suffering? Scientists sometimes argue that pure science is justified because
in the end it results in beneficial technology, but this is by no means always
the case. Research into cosmology or the unification of general relativity
and the standard model is unlikely, however successful, to have practical
consequences.
Third, there is the problem that modern science is very expensive, but
those who stand most in need of the products of science are the poor of
the earth. Ideally, science ought to serve the best interests of humanity.
That would require that the priorities of research reflect the priorities of
human need, the needs above all of the poor of the planet. It is all but inev-
itable, however, that priorities of research will reflect the interests of sci-
entists themselves, and the interests of those who pay for science: wealthy

24
This is a basic theme of Maxwell 1984: see in particular chap. 5. It is also a theme of
Kitcher 2001. Kitcher informed me that in writing his book he was influenced by his reading
of Maxwell 1984.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 335

governments, wealthy corporations, wealthy populations, and the military

of wealthy nations. The tendency of research to reflect the interests of
those who pay for it must inevitably pull priorities of research away from
responding to the most urgent and important interests of humanity. Does
not the world spend far too much on military research, and not nearly
enough on technology needed to decrease CO2 emissions rapidly, and thus
put a stop to global warming? Do the priorities of medical research reflect
the health needs of the poor in developing countries or the health needs of
those living in wealthy countries, as one might expect? These kind of con-
siderations ensure that value assumptions, and assumptions about social
use, inherent in the aims of science, will be permanently and profoundly
problematic.
Because the aims of science, in the contexts of both discovery and justi-
fication, have profoundly problematic assumptions inherent in them con-
cerning values and social use, it is, as before, vital that actual and possible
aims be imaginatively articulated and critically assessed, in a sustained
way, as an integral part of science, in an attempt to improve them. And
that means that the philosophy of science needs to be an essential, influ-
ential part of science here too.
This context of values and social use is, however, in one crucial respect,
different from contexts discussed previously. Scientists and philosophers
of science need to participate in rational—that is, imaginative and criti-
cal—exploration of actual and possible aims of science, in an attempt to
improve them. They ought not, however, to be the only participants in
this discussion. Scientists and philosophers of science cannot be in a posi-
tion to decide for the rest of humanity what is of value and what are the
world’s most important needs. Non-scientists must be able to contribute
to the discussion of aims for science on an equal footing with scientists
and philosophers of science, especially when it comes to decisions about
value and social use—decisions about priorities of research that stem from
values and social use.
The intellectual/institutional structure of science needs to be modified
to accommodate both scientists and non-scientists in the collaborative dis-
cussion of aims and priorities of scientific research. We need new journals,
new public committees, new public lectures and debates. Science journal-
ists—and philosophers of science—need to help create and keep alive this
public discussion.
Governments, funding bodies, commercial interests, charities, and sci-
entists themselves will of course continue to make decisions about what
research to support, or do. The all-important point is that this deci-
sion-making should be bathed in the results of good public exploration
of actual and possible research aims, so that decision-making may be both
improved and assessed by its means.
Throughout science, in short, in all contexts, the intellectual domain
of science needs to consist of, not just two domains, evidence and theory,

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

336 NICHOLAS MAXWELL

but as a bare minimum, three : evidence, theory, and aims. Contributions

to aims may be as important as contributions to theory or evidence. It
should be possible to win a Nobel Prize as a result of a really important
and original contribution to the domain of aims .

8. Conclusion
We philosophers of science have a major task on our hands. We need to
convince scientists—and governments and the public—that a revolution
in the nature of science is urgently needed. In a wide range of contexts,
problematic aims of science need sustained imaginative and critical dis-
cussion in an attempt so to develop science that it comes to serve the
very best interests of humanity. And furthermore, philosophers of sci-
ence need to participate in this crucial scientific activity of discussing,
and attempting to improve, problematic aims of science. Philosophers
of science need to become scientists; and some scientists at least need to
become philosophers.
Elsewhere, I have argued that there are even broader repercussions of
the above argument. I have argued that the hierarchical, aim-improving
methodology of aim-oriented empiricism has implications, when general-
ized, not just for science, but for all worthwhile human endeavours with
problematic aims.25 I have argued that it is not just science and its aims
that need to be transformed; this is true of the whole academic enterprise.
The basic intellectual aim of academic inquiry as a whole should be, not
just to acquire specialized knowledge, but rather to seek and promote
social wisdom—wisdom being the capacity, the active endeavour, and the
desire to achieve what is of value in life for oneself and others, thus includ-
ing knowledge, understanding, and technological know-how, but much
else besides.26

13 Tavistock Terrace
London, N19 4BZ
United Kingdom
[email protected]

References
Bird, Alexander. 2007. Nature’s Metaphysics: Laws and Properties.
Oxford: Clarendon Press.

25
See Maxwell 1984, chaps. 5–11; 2000; 2004, chaps. 3–4; 2014; 2017a, chap. 8; 2017c,
prologue and chap. 10; 2019b.
26
See the works referred to in the previous note.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

 A NEW TASK FOR PHILOSOPHY OF SCIENCE 337

Campbell, Joseph, Michael O’Rourke, and Matthews Slater, eds. 2011.

Carving Nature at Its Joints: Natural Kinds in Metaphysics and Science.
Cambridge, Mass.: MIT Press.
Chakravartty, Anjan. 2007. A Metaphysics for Scientific Realism.
Cambridge: Cambridge University Press.
Cory, F. J. 2000. Review of Nicholas Maxwell, The Comprehensibility of
the Universe . International Philosophical Quarterly XL, no. 4:517–18.
Dilworth, Craig. 2007. The Metaphysics of Science. Dordrecht: Springer.
Einstein, Albert. 1973. Ideas and Opinions. London: Souvenir Press.
Ellis, Brian. 2001. Scientific Essentialism. Cambridge: Cambridge
University Press.
Hossenfelder, Sabine. 2018. Lost in Math: How Beauty Leads Physics
Astray. New York: Basic Books.
Kincaid, Harold, James Ladyman, and Don Ross, eds. 2013. Scientific
Metaphysics. Oxford: Oxford University Press.
Kitcher, Philip. 1976. “Explanation, Conjunction and Unification.”
Journal of Philosophy 73:207–12.
   . 1981. “Explanatory Unification.” Philosophy of Science 48:507–31.
   . 2001. Science, Truth and Democracy. Oxford: Oxford University
Press.
Kneller, George. 1978. Science as a Human Endeavor. New York:
Columbia University Press.
Ladyman, James, Don Ross, with David Spurrett and John Collier. 2007.
Every Thing Must Go. Oxford: Oxford University Press.
Lange, Marc. 2009. Laws and Lawmakers: Science, Metaphysics and the
Laws of Nature. Oxford: Oxford University Press.
Longuet-Higgins, Christopher. 1984. “For Goodness Sake.” Nature
312:204.
Lowe, E. Jonathan. 2006. The Four-Category Ontology: A Metaphysical
Foundation for Natural Science. Oxford: Oxford University Press.
MacIntyre, Aladair. 2009. “The Very Idea of a University.” British
Journal of Educational Studies 57, no. 4:358.
Maxwell, Nicholas. 1968. “Can There Be Necessary Connections
Between Successive Events?” British Journal for the Philosophy of
Science 19, no. 1:1–25.
   . 1974. “The Rationality of Scientific Discovery: Part I.”
Philosophy of Science 41:123–53.
   . 1984. From Knowledge to Wisdom: A Revolution in the Aims and
Methods of Science. Oxford: Blackwell.
   . 1993. “Induction and Scientific Realism: Part Three: Einstein,
Aim-Oriented Empiricism and the Discovery of Special and General
Relativity.” British Journal for the Philosophy of Science 44, no.
2:275–305.
   . 1998. The Comprehensibility of the Universe: A New Conception
of Science. Oxford: Oxford University Press.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd

338 NICHOLAS MAXWELL

   . 2000. “Can Humanity Learn to Become Civilized?” Journal of

Applied Philosophy 17:2–44.
   . 2004. Is Science Neurotic? London: Imperial College Press.
   . 2005. “Popper, Kuhn, Lakatos, and Aim-Oriented Empiricism.”
Philosophia 32:181–239.
   . 2007. From Knowledge to Wisdom: A Revolution for Science and
the Humanities. 2nd extended edition. London: Pentire Press.
   . 2009. Review of Craig Dilworth, The Metaphysics of Science .
International Studies in the Philosophy of Science 23, no. 2:228–32.
   . 2013. “Has Science Established That the Cosmos Is Physically
Comprehensible?” In Recent Advances in Cosmology, edited by
Anderson Travena and Soren Brady, 1–56. New York: Nova Science.
   . 2014. How Universities Can Help Create a Wiser World. Exeter:
Imprint Academic.
   . 2017a. In Praise of Natural Philosophy. Montreal: McGill-
Queen’s University Press.
   . 2017b. Understanding Scientific Progress. St. Paul: Paragon
House.
   . 2017c. Karl Popper, Science and Enlightenment. London: UCL
Press (free online).
   . 2019a. The Metaphysics of Science and Aim-Oriented Empiricism:
A Revolution for Science and Philosophy. Cham: Springer.
   . 2019b. Science and Enlightenment: Two Great Problems of
Learning. Cham: Springer.
McHenry, Leemon. 2000. Review of Nicholas Maxwell, The Com-
prehensibility of the Universe . Mind 109:162–66.
Morganti, Matteo. 2013. Combining Science and Metaphysics. London:
Palgrave Macmillan.
Muller, F. A. 2004. “Maxwell’s Lonely War.” Studies in History and
Philosophy of Modern Physics 35:109–10, and 117.
Mumford, Stephen, and Matthews Tugby, eds. 2013. Metaphysics and
Science. Oxford: Oxford University Press.
Popper, Karl. 1959. The Logic of Scientific Discovery. London:
Hutchinson.
Roush, Sherrilyn. 2001. Review of Nicholas Maxwell, The Com-
prehensibility of the Universe . Philosophical Review 110:85–87.
Slater, Matthew, and Zanjas Yudell, eds. 2017. Metaphysics and the
Philosophy of Science. Oxford: Oxford University Press.
Smart, John Jamieson Carswell. 2000. Review of Nicholas Maxwell, The
Comprehensibility of the Universe . British Journal for the Philosophy of
Science 51:907–11.
Trout, J. D. 2016. Wondrous Truths. Oxford: Oxford University Press.

© 2019 Metaphilosophy LLC and John Wiley & Sons Ltd